* Copyright (C) 1999 Erik Walthinsen <omega@cse.ogi.edu>
* Copyright (C) 2004 Wim Taymans <wim.taymans@gmail.com>
* Copyright (C) 2007 Peter Kjellerstedt <pkj@axis.com>
- * Copyright (C) 2008 Ole André Vadla Ravnås <ole.andre.ravnas@tandberg.com>
+ * Copyright (C) 2008 Ole André Vadla Ravnås <ole.andre.ravnas@tandberg.com>
*
* gstpoll.c: File descriptor set
*
*
* You should have received a copy of the GNU Library General Public
* License along with this library; if not, write to the
- * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
- * Boston, MA 02111-1307, USA.
+ * Free Software Foundation, Inc., 51 Franklin St, Fifth Floor,
+ * Boston, MA 02110-1301, USA.
*/
/**
* SECTION:gstpoll
+ * @title: GstPoll
* @short_description: Keep track of file descriptors and make it possible
- * to wait on them in a cancelable way
+ * to wait on them in a cancellable way
*
* A #GstPoll keeps track of file descriptors much like fd_set (used with
* select()) or a struct pollfd array (used with poll()). Once created with
* gst_poll_new(), the set can be used to wait for file descriptors to be
- * readable and/or writeable. It is possible to make this wait be controlled
+ * readable and/or writable. It is possible to make this wait be controlled
* by specifying %TRUE for the @controllable flag when creating the set (or
* later calling gst_poll_set_controllable()).
*
* New file descriptors are added to the set using gst_poll_add_fd(), and
* removed using gst_poll_remove_fd(). Controlling which file descriptors
- * should be waited for to become readable and/or writeable are done using
- * gst_poll_fd_ctl_read() and gst_poll_fd_ctl_write().
+ * should be waited for to become readable and/or writable are done using
+ * gst_poll_fd_ctl_read(), gst_poll_fd_ctl_write() and gst_poll_fd_ctl_pri().
*
* Use gst_poll_wait() to wait for the file descriptors to actually become
- * readable and/or writeable, or to timeout if no file descriptor is available
+ * readable and/or writable, or to timeout if no file descriptor is available
* in time. The wait can be controlled by calling gst_poll_restart() and
* gst_poll_set_flushing().
*
#include "config.h"
#endif
-#ifndef _MSC_VER
-#define _GNU_SOURCE 1
-#include <sys/poll.h>
-#include <sys/time.h>
-#endif
+#include "gst_private.h"
+#include "glib-compat-private.h"
#include <sys/types.h>
#ifdef G_OS_WIN32
#include <winsock2.h>
-#define EINPROGRESS WSAEINPROGRESS
#else
+#define _GNU_SOURCE 1
+#ifdef HAVE_SYS_POLL_H
+#include <sys/poll.h>
+#endif
+#ifdef HAVE_POLL_H
+#include <poll.h>
+#endif
+#include <sys/time.h>
#include <sys/socket.h>
#endif
+#ifdef G_OS_WIN32
+# ifndef EWOULDBLOCK
+# define EWOULDBLOCK EAGAIN /* This is just to placate gcc */
+# endif
+#endif /* G_OS_WIN32 */
+
/* OS/X needs this because of bad headers */
#include <string.h>
-#include "gst_private.h"
-
-#include "gstpoll.h"
-
-#ifndef G_OS_WIN32
-/* the poll/select call is also performed on a control socket, that way
- * we can send special commands to control it
+/* The poll() emulation on OS/X doesn't handle fds=NULL, nfds=0,
+ * so we prefer our own poll emulation.
*/
-#define SEND_COMMAND(set, command) \
-G_STMT_START { \
- unsigned char c = command; \
- write (set->control_write_fd.fd, &c, 1); \
-} G_STMT_END
+#if defined(BROKEN_POLL)
+#undef HAVE_POLL
+#endif
-#define READ_COMMAND(set, command, res) \
-G_STMT_START { \
- res = read (set->control_read_fd.fd, &command, 1); \
-} G_STMT_END
+#include "gstpoll.h"
-#define GST_POLL_CMD_WAKEUP 'W' /* restart the poll/select call */
+#define GST_CAT_DEFAULT GST_CAT_POLL
-#else /* G_OS_WIN32 */
+#ifdef G_OS_WIN32
typedef struct _WinsockFd WinsockFd;
struct _WinsockFd
{
GstPollMode mode;
- GMutex *lock;
-
+ GMutex lock;
+ /* array of fds, always written to and read from with lock */
GArray *fds;
+ /* array of active fds, only written to from the waiting thread with the
+ * lock and read from with the lock or without the lock from the waiting
+ * thread */
GArray *active_fds;
+
#ifndef G_OS_WIN32
GstPollFD control_read_fd;
GstPollFD control_write_fd;
#else
GArray *active_fds_ignored;
-
GArray *events;
GArray *active_events;
#endif
gboolean controllable;
- gboolean new_controllable;
- gboolean waiting;
- gboolean flushing;
+ volatile gint waiting;
+ volatile gint control_pending;
+ volatile gint flushing;
+ gboolean timer;
+ volatile gint rebuild;
};
+static gboolean gst_poll_fd_ctl_read_unlocked (GstPoll * set, GstPollFD * fd,
+ gboolean active);
+static gboolean gst_poll_add_fd_unlocked (GstPoll * set, GstPollFD * fd);
+
+#define IS_FLUSHING(s) (g_atomic_int_get(&(s)->flushing))
+#define SET_FLUSHING(s,val) (g_atomic_int_set(&(s)->flushing, (val)))
+
+#define INC_WAITING(s) (g_atomic_int_add(&(s)->waiting, 1))
+#define DEC_WAITING(s) (g_atomic_int_add(&(s)->waiting, -1))
+#define GET_WAITING(s) (g_atomic_int_get(&(s)->waiting))
+
+#define TEST_REBUILD(s) (g_atomic_int_compare_and_exchange(&(s)->rebuild, 1, 0))
+#define MARK_REBUILD(s) (g_atomic_int_set(&(s)->rebuild, 1))
+
+#ifndef G_OS_WIN32
+
+static gboolean
+wake_event (GstPoll * set)
+{
+ ssize_t num_written;
+ while ((num_written = write (set->control_write_fd.fd, "W", 1)) != 1) {
+ if (num_written == -1 && errno != EAGAIN && errno != EINTR) {
+ g_critical ("%p: failed to wake event: %s", set, strerror (errno));
+ return FALSE;
+ }
+ }
+ return TRUE;
+}
+
+static gboolean
+release_event (GstPoll * set)
+{
+ gchar buf[1] = { '\0' };
+ ssize_t num_read;
+ while ((num_read = read (set->control_read_fd.fd, buf, 1)) != 1) {
+ if (num_read == -1 && errno != EAGAIN && errno != EINTR) {
+ g_critical ("%p: failed to release event: %s", set, strerror (errno));
+ return FALSE;
+ }
+ }
+ return TRUE;
+}
+
+#else
+
+static void
+format_last_error (gchar * buf, size_t buf_len)
+{
+ DWORD flags = FORMAT_MESSAGE_FROM_SYSTEM;
+ LPCVOID src = NULL;
+ DWORD lang = 0;
+ DWORD id;
+ id = GetLastError ();
+ FormatMessage (flags, src, id, lang, buf, (DWORD) buf_len, NULL);
+ SetLastError (id);
+}
+
+static gboolean
+wake_event (GstPoll * set)
+{
+ SetLastError (0);
+ errno = 0;
+ if (!SetEvent (set->wakeup_event)) {
+ gchar msg[1024] = "<unknown>";
+ format_last_error (msg, sizeof (msg));
+ g_critical ("%p: failed to set wakup_event: %s", set, msg);
+ errno = EBADF;
+ return FALSE;
+ }
+
+ return TRUE;
+}
+
+static gboolean
+release_event (GstPoll * set)
+{
+ DWORD status;
+ SetLastError (0);
+ errno = 0;
+
+ status = WaitForSingleObject (set->wakeup_event, INFINITE);
+ if (status) {
+ const gchar *reason = "unknown";
+ gchar msg[1024] = "<unknown>";
+ switch (status) {
+ case WAIT_ABANDONED:
+ reason = "WAIT_ABANDONED";
+ break;
+ case WAIT_TIMEOUT:
+ reason = "WAIT_TIMEOUT";
+ break;
+ case WAIT_FAILED:
+ format_last_error (msg, sizeof (msg));
+ reason = msg;
+ break;
+ default:
+ reason = "other";
+ break;
+ }
+ g_critical ("%p: failed to block on wakup_event: %s", set, reason);
+ errno = EBADF;
+ return FALSE;
+ }
+
+ if (!ResetEvent (set->wakeup_event)) {
+ gchar msg[1024] = "<unknown>";
+ format_last_error (msg, sizeof (msg));
+ g_critical ("%p: failed to reset wakup_event: %s", set, msg);
+ errno = EBADF;
+ return FALSE;
+ }
+
+ return TRUE;
+}
+
+#endif
+
+/* the poll/select call is also performed on a control socket, that way
+ * we can send special commands to control it */
+static inline gboolean
+raise_wakeup (GstPoll * set)
+{
+ gboolean result = TRUE;
+
+ /* makes testing control_pending and WAKE_EVENT() atomic. */
+ g_mutex_lock (&set->lock);
+
+ if (set->control_pending == 0) {
+ /* raise when nothing pending */
+ GST_LOG ("%p: raise", set);
+ result = wake_event (set);
+ }
+
+ if (result) {
+ set->control_pending++;
+ }
+
+ g_mutex_unlock (&set->lock);
+
+ return result;
+}
+
+static inline gboolean
+release_wakeup (GstPoll * set)
+{
+ gboolean result = FALSE;
+
+ /* makes testing/modifying control_pending and RELEASE_EVENT() atomic. */
+ g_mutex_lock (&set->lock);
+
+ if (set->control_pending > 0) {
+ /* release, only if this was the last pending. */
+ if (set->control_pending == 1) {
+ GST_LOG ("%p: release", set);
+ result = release_event (set);
+ } else {
+ result = TRUE;
+ }
+
+ if (result) {
+ set->control_pending--;
+ }
+ } else {
+ errno = EWOULDBLOCK;
+ }
+
+ g_mutex_unlock (&set->lock);
+
+ return result;
+}
+
+static inline gint
+release_all_wakeup (GstPoll * set)
+{
+ gint old;
+
+ /* makes testing control_pending and RELEASE_EVENT() atomic. */
+ g_mutex_lock (&set->lock);
+
+ if ((old = set->control_pending) > 0) {
+ GST_LOG ("%p: releasing %d", set, old);
+ if (release_event (set)) {
+ set->control_pending = 0;
+ } else {
+ old = 0;
+ }
+ }
+
+ g_mutex_unlock (&set->lock);
+
+ return old;
+}
+
static gint
find_index (GArray * array, GstPollFD * fd)
{
#if !defined(HAVE_PPOLL) && defined(HAVE_POLL)
/* check if all file descriptors will fit in an fd_set */
static gboolean
-selectable_fds (const GstPoll * set)
+selectable_fds (GstPoll * set)
{
guint i;
+ g_mutex_lock (&set->lock);
for (i = 0; i < set->fds->len; i++) {
struct pollfd *pfd = &g_array_index (set->fds, struct pollfd, i);
if (pfd->fd >= FD_SETSIZE)
- return FALSE;
+ goto too_many;
}
+ g_mutex_unlock (&set->lock);
return TRUE;
+
+too_many:
+ {
+ g_mutex_unlock (&set->lock);
+ return FALSE;
+ }
}
/* check if the timeout will convert to a timeout value used for poll()
#endif
static GstPollMode
-choose_mode (const GstPoll * set, GstClockTime timeout)
+choose_mode (GstPoll * set, GstClockTime timeout)
{
GstPollMode mode;
#ifndef G_OS_WIN32
static gint
-pollfd_to_fd_set (GstPoll * set, fd_set * readfds, fd_set * writefds)
+pollfd_to_fd_set (GstPoll * set, fd_set * readfds, fd_set * writefds,
+ fd_set * errorfds)
{
gint max_fd = -1;
guint i;
FD_ZERO (readfds);
FD_ZERO (writefds);
+ FD_ZERO (errorfds);
- g_mutex_lock (set->lock);
+ g_mutex_lock (&set->lock);
for (i = 0; i < set->active_fds->len; i++) {
struct pollfd *pfd = &g_array_index (set->fds, struct pollfd, i);
FD_SET (pfd->fd, readfds);
if (pfd->events & POLLOUT)
FD_SET (pfd->fd, writefds);
- if (pfd->fd > max_fd)
+ if (pfd->events)
+ FD_SET (pfd->fd, errorfds);
+ if (pfd->fd > max_fd && (pfd->events & (POLLIN | POLLOUT)))
max_fd = pfd->fd;
}
}
- g_mutex_unlock (set->lock);
+ g_mutex_unlock (&set->lock);
return max_fd;
}
static void
-fd_set_to_pollfd (GstPoll * set, fd_set * readfds, fd_set * writefds)
+fd_set_to_pollfd (GstPoll * set, fd_set * readfds, fd_set * writefds,
+ fd_set * errorfds)
{
guint i;
- g_mutex_lock (set->lock);
+ g_mutex_lock (&set->lock);
for (i = 0; i < set->active_fds->len; i++) {
struct pollfd *pfd = &g_array_index (set->active_fds, struct pollfd, i);
if (pfd->fd < FD_SETSIZE) {
+ pfd->revents = 0;
if (FD_ISSET (pfd->fd, readfds))
pfd->revents |= POLLIN;
if (FD_ISSET (pfd->fd, writefds))
pfd->revents |= POLLOUT;
+ if (FD_ISSET (pfd->fd, errorfds))
+ pfd->revents |= POLLERR;
}
}
- g_mutex_unlock (set->lock);
+ g_mutex_unlock (&set->lock);
}
#else /* G_OS_WIN32 */
/*
#endif
/**
- * gst_poll_new:
+ * gst_poll_new: (skip)
* @controllable: whether it should be possible to control a wait.
*
* Create a new file descriptor set. If @controllable, it
* is possible to restart or flush a call to gst_poll_wait() with
* gst_poll_restart() and gst_poll_set_flushing() respectively.
*
- * Returns: a new #GstPoll, or %NULL in case of an error. Free with
- * gst_poll_free().
+ * Free-function: gst_poll_free
*
- * Since: 0.10.18
+ * Returns: (transfer full) (nullable): a new #GstPoll, or %NULL in
+ * case of an error. Free with gst_poll_free().
*/
GstPoll *
gst_poll_new (gboolean controllable)
{
GstPoll *nset;
- nset = g_new0 (GstPoll, 1);
- nset->lock = g_mutex_new ();
+ nset = g_slice_new0 (GstPoll);
+ GST_DEBUG ("%p: new controllable : %d", nset, controllable);
+ g_mutex_init (&nset->lock);
#ifndef G_OS_WIN32
nset->mode = GST_POLL_MODE_AUTO;
nset->fds = g_array_new (FALSE, FALSE, sizeof (struct pollfd));
nset->active_fds = g_array_new (FALSE, FALSE, sizeof (struct pollfd));
nset->control_read_fd.fd = -1;
nset->control_write_fd.fd = -1;
+ {
+ gint control_sock[2];
+
+ if (socketpair (PF_UNIX, SOCK_STREAM, 0, control_sock) < 0)
+ goto no_socket_pair;
+
+ nset->control_read_fd.fd = control_sock[0];
+ nset->control_write_fd.fd = control_sock[1];
+
+ gst_poll_add_fd_unlocked (nset, &nset->control_read_fd);
+ gst_poll_fd_ctl_read_unlocked (nset, &nset->control_read_fd, TRUE);
+ }
#else
nset->mode = GST_POLL_MODE_WINDOWS;
nset->fds = g_array_new (FALSE, FALSE, sizeof (WinsockFd));
nset->wakeup_event = CreateEvent (NULL, TRUE, FALSE, NULL);
#endif
- if (!gst_poll_set_controllable (nset, controllable))
- goto not_controllable;
+ /* ensure (re)build, though already sneakily set in non-windows case */
+ MARK_REBUILD (nset);
+
+ nset->controllable = controllable;
+ nset->control_pending = 0;
return nset;
/* ERRORS */
-not_controllable:
+#ifndef G_OS_WIN32
+no_socket_pair:
{
+ GST_WARNING ("%p: can't create socket pair !", nset);
gst_poll_free (nset);
return NULL;
}
+#endif
+}
+
+/**
+ * gst_poll_new_timer: (skip)
+ *
+ * Create a new poll object that can be used for scheduling cancellable
+ * timeouts.
+ *
+ * A timeout is performed with gst_poll_wait(). Multiple timeouts can be
+ * performed from different threads.
+ *
+ * Free-function: gst_poll_free
+ *
+ * Returns: (transfer full) (nullable): a new #GstPoll, or %NULL in
+ * case of an error. Free with gst_poll_free().
+ */
+GstPoll *
+gst_poll_new_timer (void)
+{
+ GstPoll *poll;
+
+ /* make a new controllable poll set */
+ if (!(poll = gst_poll_new (TRUE)))
+ goto done;
+
+ /* we are a timer */
+ poll->timer = TRUE;
+
+done:
+ return poll;
}
/**
* gst_poll_free:
- * @set: a file descriptor set.
+ * @set: (transfer full): a file descriptor set.
*
* Free a file descriptor set.
- *
- * Since: 0.10.18
*/
void
gst_poll_free (GstPoll * set)
{
g_return_if_fail (set != NULL);
+ GST_DEBUG ("%p: freeing", set);
+
#ifndef G_OS_WIN32
if (set->control_write_fd.fd >= 0)
close (set->control_write_fd.fd);
g_array_free (set->active_fds, TRUE);
g_array_free (set->fds, TRUE);
- g_mutex_free (set->lock);
- g_free (set);
+ g_mutex_clear (&set->lock);
+ g_slice_free (GstPoll, set);
+}
+
+/**
+ * gst_poll_get_read_gpollfd:
+ * @set: a #GstPoll
+ * @fd: a #GPollFD
+ *
+ * Get a GPollFD for the reading part of the control socket. This is useful when
+ * integrating with a GSource and GMainLoop.
+ */
+void
+gst_poll_get_read_gpollfd (GstPoll * set, GPollFD * fd)
+{
+ g_return_if_fail (set != NULL);
+ g_return_if_fail (fd != NULL);
+
+#ifndef G_OS_WIN32
+ fd->fd = set->control_read_fd.fd;
+#else
+#if GLIB_SIZEOF_VOID_P == 8
+ fd->fd = (gint64) set->wakeup_event;
+#else
+ fd->fd = (gint) set->wakeup_event;
+#endif
+#endif
+ fd->events = G_IO_IN | G_IO_HUP | G_IO_ERR;
+ fd->revents = 0;
}
/**
*
* Initializes @fd. Alternatively you can initialize it with
* #GST_POLL_FD_INIT.
- *
- * Since: 0.10.18
*/
void
gst_poll_fd_init (GstPollFD * fd)
{
gint idx;
+ GST_DEBUG ("%p: fd (fd:%d, idx:%d)", set, fd->fd, fd->idx);
+
idx = find_index (set->fds, fd);
if (idx < 0) {
#ifndef G_OS_WIN32
fd->idx = set->fds->len - 1;
#endif
+ MARK_REBUILD (set);
+ } else {
+ GST_WARNING ("%p: fd already added !", set);
}
return TRUE;
* Add a file descriptor to the file descriptor set.
*
* Returns: %TRUE if the file descriptor was successfully added to the set.
- *
- * Since: 0.10.18
*/
gboolean
gst_poll_add_fd (GstPoll * set, GstPollFD * fd)
g_return_val_if_fail (fd != NULL, FALSE);
g_return_val_if_fail (fd->fd >= 0, FALSE);
- g_mutex_lock (set->lock);
+ g_mutex_lock (&set->lock);
ret = gst_poll_add_fd_unlocked (set, fd);
- g_mutex_unlock (set->lock);
+ g_mutex_unlock (&set->lock);
return ret;
}
* Remove a file descriptor from the file descriptor set.
*
* Returns: %TRUE if the file descriptor was successfully removed from the set.
- *
- * Since: 0.10.18
*/
gboolean
gst_poll_remove_fd (GstPoll * set, GstPollFD * fd)
g_return_val_if_fail (fd != NULL, FALSE);
g_return_val_if_fail (fd->fd >= 0, FALSE);
- g_mutex_lock (set->lock);
+
+ GST_DEBUG ("%p: fd (fd:%d, idx:%d)", set, fd->fd, fd->idx);
+
+ g_mutex_lock (&set->lock);
/* get the index, -1 is an fd that is not added */
idx = find_index (set->fds, fd);
/* mark fd as removed by setting the index to -1 */
fd->idx = -1;
+ MARK_REBUILD (set);
+ } else {
+ GST_WARNING ("%p: couldn't find fd !", set);
}
- g_mutex_unlock (set->lock);
+ g_mutex_unlock (&set->lock);
return idx >= 0;
}
* writability.
*
* Returns: %TRUE if the descriptor was successfully updated.
- *
- * Since: 0.10.18
*/
gboolean
gst_poll_fd_ctl_write (GstPoll * set, GstPollFD * fd, gboolean active)
g_return_val_if_fail (fd != NULL, FALSE);
g_return_val_if_fail (fd->fd >= 0, FALSE);
- g_mutex_lock (set->lock);
+ GST_DEBUG ("%p: fd (fd:%d, idx:%d), active : %d", set,
+ fd->fd, fd->idx, active);
+
+ g_mutex_lock (&set->lock);
idx = find_index (set->fds, fd);
if (idx >= 0) {
pfd->events |= POLLOUT;
else
pfd->events &= ~POLLOUT;
+
+ GST_LOG ("%p: pfd->events now %d (POLLOUT:%d)", set, pfd->events, POLLOUT);
#else
- gst_poll_update_winsock_event_mask (set, idx, FD_WRITE, active);
+ gst_poll_update_winsock_event_mask (set, idx, FD_WRITE | FD_CONNECT,
+ active);
#endif
+ MARK_REBUILD (set);
+ } else {
+ GST_WARNING ("%p: couldn't find fd !", set);
}
- g_mutex_unlock (set->lock);
+ g_mutex_unlock (&set->lock);
return idx >= 0;
}
{
gint idx;
+ GST_DEBUG ("%p: fd (fd:%d, idx:%d), active : %d", set,
+ fd->fd, fd->idx, active);
+
idx = find_index (set->fds, fd);
if (idx >= 0) {
struct pollfd *pfd = &g_array_index (set->fds, struct pollfd, idx);
if (active)
- pfd->events |= (POLLIN | POLLPRI);
+ pfd->events |= POLLIN;
else
- pfd->events &= ~(POLLIN | POLLPRI);
+ pfd->events &= ~POLLIN;
#else
gst_poll_update_winsock_event_mask (set, idx, FD_READ | FD_ACCEPT, active);
#endif
+ MARK_REBUILD (set);
+ } else {
+ GST_WARNING ("%p: couldn't find fd !", set);
}
return idx >= 0;
* readability.
*
* Returns: %TRUE if the descriptor was successfully updated.
- *
- * Since: 0.10.18
*/
gboolean
gst_poll_fd_ctl_read (GstPoll * set, GstPollFD * fd, gboolean active)
g_return_val_if_fail (fd != NULL, FALSE);
g_return_val_if_fail (fd->fd >= 0, FALSE);
- g_mutex_lock (set->lock);
+ g_mutex_lock (&set->lock);
ret = gst_poll_fd_ctl_read_unlocked (set, fd, active);
- g_mutex_unlock (set->lock);
+ g_mutex_unlock (&set->lock);
return ret;
}
/**
+ * gst_poll_fd_ctl_pri:
+ * @set: a file descriptor set.
+ * @fd: a file descriptor.
+ * @active: a new status.
+ *
+ * Control whether the descriptor @fd in @set will be monitored for
+ * exceptional conditions (POLLPRI).
+ *
+ * Not implemented on Windows (will just return %FALSE there).
+ *
+ * Returns: %TRUE if the descriptor was successfully updated.
+ *
+ * Since: 1.16
+ */
+gboolean
+gst_poll_fd_ctl_pri (GstPoll * set, GstPollFD * fd, gboolean active)
+{
+#ifdef G_OS_WIN32
+ return FALSE;
+#else
+ gint idx;
+
+ g_return_val_if_fail (set != NULL, FALSE);
+ g_return_val_if_fail (fd != NULL, FALSE);
+ g_return_val_if_fail (fd->fd >= 0, FALSE);
+
+ GST_DEBUG ("%p: fd (fd:%d, idx:%d), active : %d", set,
+ fd->fd, fd->idx, active);
+
+ g_mutex_lock (&set->lock);
+
+ idx = find_index (set->fds, fd);
+ if (idx >= 0) {
+ struct pollfd *pfd = &g_array_index (set->fds, struct pollfd, idx);
+
+ if (active)
+ pfd->events |= POLLPRI;
+ else
+ pfd->events &= ~POLLPRI;
+
+ GST_LOG ("%p: pfd->events now %d (POLLPRI:%d)", set, pfd->events, POLLOUT);
+ MARK_REBUILD (set);
+ } else {
+ GST_WARNING ("%p: couldn't find fd !", set);
+ }
+
+ g_mutex_unlock (&set->lock);
+
+ return idx >= 0;
+#endif
+}
+
+/**
* gst_poll_fd_ignored:
* @set: a file descriptor set.
* @fd: a file descriptor.
* The reason why this is needed is because the underlying implementation
* might not allow querying the fd more than once between calls to one of
* the re-enabling operations.
- *
- * Since: 0.10.18
*/
void
gst_poll_fd_ignored (GstPoll * set, GstPollFD * fd)
g_return_if_fail (fd != NULL);
g_return_if_fail (fd->fd >= 0);
- g_mutex_lock (set->lock);
+ g_mutex_lock (&set->lock);
idx = find_index (set->fds, fd);
if (idx >= 0) {
WinsockFd *wfd = &g_array_index (set->fds, WinsockFd, idx);
wfd->ignored_event_mask = wfd->event_mask & (FD_READ | FD_WRITE);
+ MARK_REBUILD (set);
}
- g_mutex_unlock (set->lock);
+ g_mutex_unlock (&set->lock);
#endif
}
* Check if @fd in @set has closed the connection.
*
* Returns: %TRUE if the connection was closed.
- *
- * Since: 0.10.18
*/
gboolean
gst_poll_fd_has_closed (const GstPoll * set, GstPollFD * fd)
g_return_val_if_fail (fd != NULL, FALSE);
g_return_val_if_fail (fd->fd >= 0, FALSE);
- g_mutex_lock (set->lock);
+ g_mutex_lock (&((GstPoll *) set)->lock);
idx = find_index (set->active_fds, fd);
if (idx >= 0) {
res = (wfd->events.lNetworkEvents & FD_CLOSE) != 0;
#endif
+ } else {
+ GST_WARNING ("%p: couldn't find fd !", set);
}
+ g_mutex_unlock (&((GstPoll *) set)->lock);
- g_mutex_unlock (set->lock);
+ GST_DEBUG ("%p: fd (fd:%d, idx:%d) %d", set, fd->fd, fd->idx, res);
return res;
}
* Check if @fd in @set has an error.
*
* Returns: %TRUE if the descriptor has an error.
- *
- * Since: 0.10.18
*/
gboolean
gst_poll_fd_has_error (const GstPoll * set, GstPollFD * fd)
g_return_val_if_fail (fd != NULL, FALSE);
g_return_val_if_fail (fd->fd >= 0, FALSE);
- g_mutex_lock (set->lock);
+ g_mutex_lock (&((GstPoll *) set)->lock);
idx = find_index (set->active_fds, fd);
if (idx >= 0) {
res = (wfd->events.iErrorCode[FD_CLOSE_BIT] != 0) ||
(wfd->events.iErrorCode[FD_READ_BIT] != 0) ||
(wfd->events.iErrorCode[FD_WRITE_BIT] != 0) ||
- (wfd->events.iErrorCode[FD_ACCEPT_BIT] != 0);
+ (wfd->events.iErrorCode[FD_ACCEPT_BIT] != 0) ||
+ (wfd->events.iErrorCode[FD_CONNECT_BIT] != 0);
#endif
+ } else {
+ GST_WARNING ("%p: couldn't find fd !", set);
}
+ g_mutex_unlock (&((GstPoll *) set)->lock);
- g_mutex_unlock (set->lock);
+ GST_DEBUG ("%p: fd (fd:%d, idx:%d) %d", set, fd->fd, fd->idx, res);
return res;
}
#ifndef G_OS_WIN32
struct pollfd *pfd = &g_array_index (set->active_fds, struct pollfd, idx);
- res = (pfd->revents & (POLLIN | POLLPRI)) != 0;
+ res = (pfd->revents & POLLIN) != 0;
#else
WinsockFd *wfd = &g_array_index (set->active_fds, WinsockFd, idx);
res = (wfd->events.lNetworkEvents & (FD_READ | FD_ACCEPT)) != 0;
#endif
+ } else {
+ GST_WARNING ("%p: couldn't find fd !", set);
}
+ GST_DEBUG ("%p: fd (fd:%d, idx:%d) %d", set, fd->fd, fd->idx, res);
return res;
}
* Check if @fd in @set has data to be read.
*
* Returns: %TRUE if the descriptor has data to be read.
- *
- * Since: 0.10.18
*/
gboolean
gst_poll_fd_can_read (const GstPoll * set, GstPollFD * fd)
g_return_val_if_fail (fd != NULL, FALSE);
g_return_val_if_fail (fd->fd >= 0, FALSE);
- g_mutex_lock (set->lock);
+ g_mutex_lock (&((GstPoll *) set)->lock);
res = gst_poll_fd_can_read_unlocked (set, fd);
- g_mutex_unlock (set->lock);
+ g_mutex_unlock (&((GstPoll *) set)->lock);
return res;
}
* Check if @fd in @set can be used for writing.
*
* Returns: %TRUE if the descriptor can be used for writing.
- *
- * Since: 0.10.18
*/
gboolean
gst_poll_fd_can_write (const GstPoll * set, GstPollFD * fd)
g_return_val_if_fail (fd != NULL, FALSE);
g_return_val_if_fail (fd->fd >= 0, FALSE);
- g_mutex_lock (set->lock);
+ g_mutex_lock (&((GstPoll *) set)->lock);
idx = find_index (set->active_fds, fd);
if (idx >= 0) {
res = (wfd->events.lNetworkEvents & FD_WRITE) != 0;
#endif
+ } else {
+ GST_WARNING ("%p: couldn't find fd !", set);
}
+ g_mutex_unlock (&((GstPoll *) set)->lock);
- g_mutex_unlock (set->lock);
+ GST_DEBUG ("%p: fd (fd:%d, idx:%d) %d", set, fd->fd, fd->idx, res);
return res;
}
-static void
-gst_poll_check_ctrl_commands (GstPoll * set, gint res, gboolean * restarting)
+/**
+ * gst_poll_fd_has_pri:
+ * @set: a file descriptor set.
+ * @fd: a file descriptor.
+ *
+ * Check if @fd in @set has an exceptional condition (POLLPRI).
+ *
+ * Not implemented on Windows (will just return %FALSE there).
+ *
+ * Returns: %TRUE if the descriptor has an exceptional condition.
+ *
+ * Since: 1.16
+ */
+gboolean
+gst_poll_fd_has_pri (const GstPoll * set, GstPollFD * fd)
{
- /* check if the poll/select was aborted due to a command */
- if (set->controllable) {
-#ifndef G_OS_WIN32
- while (TRUE) {
- guchar cmd;
- gint result;
-
- /* we do not check the read status of the control socket here because
- * there may have been a write to the socket between the time the
- * poll/select finished and before we got the mutex back, and we need
- * to clear out the control socket before leaving */
- READ_COMMAND (set, cmd, result);
- if (result <= 0) {
- /* no more commands, quit the loop */
- break;
- }
-
- /* if the control socket is the only socket with activity when we get
- * here, we restart the _wait operation, else we allow the caller to
- * process the other file descriptors */
- if (res == 1 &&
- gst_poll_fd_can_read_unlocked (set, &set->control_read_fd))
- *restarting = TRUE;
- }
+#ifdef G_OS_WIN32
+ return FALSE;
#else
- if (WaitForSingleObject (set->wakeup_event, 0) == WAIT_OBJECT_0) {
- ResetEvent (set->wakeup_event);
- *restarting = TRUE;
- }
-#endif
+ gboolean res = FALSE;
+ gint idx;
+
+ g_return_val_if_fail (set != NULL, FALSE);
+ g_return_val_if_fail (fd != NULL, FALSE);
+ g_return_val_if_fail (fd->fd >= 0, FALSE);
+
+ g_mutex_lock (&((GstPoll *) set)->lock);
+
+ idx = find_index (set->active_fds, fd);
+ if (idx >= 0) {
+ struct pollfd *pfd = &g_array_index (set->active_fds, struct pollfd, idx);
+
+ res = (pfd->revents & POLLPRI) != 0;
+ } else {
+ GST_WARNING ("%p: couldn't find fd !", set);
}
+ g_mutex_unlock (&((GstPoll *) set)->lock);
+
+ GST_DEBUG ("%p: fd (fd:%d, idx:%d) %d", set, fd->fd, fd->idx, res);
+
+ return res;
+#endif
}
/**
* Wait for activity on the file descriptors in @set. This function waits up to
* the specified @timeout. A timeout of #GST_CLOCK_TIME_NONE waits forever.
*
- * When this function is called from multiple threads, -1 will be returned with
- * errno set to EPERM.
+ * For #GstPoll objects created with gst_poll_new(), this function can only be
+ * called from a single thread at a time. If called from multiple threads,
+ * -1 will be returned with errno set to EPERM.
+ *
+ * This is not true for timer #GstPoll objects created with
+ * gst_poll_new_timer(), where it is allowed to have multiple threads waiting
+ * simultaneously.
*
* Returns: The number of #GstPollFD in @set that have activity or 0 when no
* activity was detected after @timeout. If an error occurs, -1 is returned
* and errno is set.
- *
- * Since: 0.10.18
*/
gint
gst_poll_wait (GstPoll * set, GstClockTime timeout)
{
gboolean restarting;
+ gboolean is_timer;
int res;
+ gint old_waiting;
g_return_val_if_fail (set != NULL, -1);
- g_mutex_lock (set->lock);
+ GST_DEBUG ("%p: timeout :%" GST_TIME_FORMAT, set, GST_TIME_ARGS (timeout));
+
+ is_timer = set->timer;
- /* we cannot wait from multiple threads */
- if (set->waiting)
+ /* add one more waiter */
+ old_waiting = INC_WAITING (set);
+
+ /* we cannot wait from multiple threads unless we are a timer */
+ if (G_UNLIKELY (old_waiting > 0 && !is_timer))
goto already_waiting;
- /* flushing, exit immediatly */
- if (set->flushing)
+ /* flushing, exit immediately */
+ if (G_UNLIKELY (IS_FLUSHING (set)))
goto flushing;
- set->waiting = TRUE;
-
do {
GstPollMode mode;
mode = choose_mode (set, timeout);
+ if (TEST_REBUILD (set)) {
+ g_mutex_lock (&set->lock);
#ifndef G_OS_WIN32
- g_array_set_size (set->active_fds, set->fds->len);
- memcpy (set->active_fds->data, set->fds->data,
- set->fds->len * sizeof (struct pollfd));
+ g_array_set_size (set->active_fds, set->fds->len);
+ memcpy (set->active_fds->data, set->fds->data,
+ set->fds->len * sizeof (struct pollfd));
#else
- if (!gst_poll_prepare_winsock_active_sets (set))
- goto winsock_error;
+ if (!gst_poll_prepare_winsock_active_sets (set))
+ goto winsock_error;
#endif
-
- g_mutex_unlock (set->lock);
+ g_mutex_unlock (&set->lock);
+ }
switch (mode) {
case GST_POLL_MODE_AUTO:
#ifndef G_OS_WIN32
fd_set readfds;
fd_set writefds;
+ fd_set errorfds;
gint max_fd;
- max_fd = pollfd_to_fd_set (set, &readfds, &writefds);
+ max_fd = pollfd_to_fd_set (set, &readfds, &writefds, &errorfds);
if (mode == GST_POLL_MODE_SELECT) {
struct timeval tv;
tvptr = NULL;
}
- res = select (max_fd + 1, &readfds, &writefds, NULL, tvptr);
+ GST_DEBUG ("%p: Calling select", set);
+ res = select (max_fd + 1, &readfds, &writefds, &errorfds, tvptr);
+ GST_DEBUG ("%p: After select, res:%d", set, res);
} else {
#ifdef HAVE_PSELECT
struct timespec ts;
tsptr = NULL;
}
- res = pselect (max_fd + 1, &readfds, &writefds, NULL, tsptr, NULL);
+ GST_DEBUG ("%p: Calling pselect", set);
+ res =
+ pselect (max_fd + 1, &readfds, &writefds, &errorfds, tsptr, NULL);
+ GST_DEBUG ("%p: After pselect, res:%d", set, res);
#endif
}
- if (res > 0) {
- fd_set_to_pollfd (set, &readfds, &writefds);
+ if (res >= 0) {
+ fd_set_to_pollfd (set, &readfds, &writefds, &errorfds);
}
#else /* G_OS_WIN32 */
g_assert_not_reached ();
t = 0;
}
- wait_ret = WSAWaitForMultipleEvents (set->active_events->len,
- (HANDLE *) set->active_events->data, FALSE, t, FALSE);
+ if (set->active_events->len != 0) {
+ wait_ret = WSAWaitForMultipleEvents (set->active_events->len,
+ (HANDLE *) set->active_events->data, FALSE, t, FALSE);
+ } else {
+ wait_ret = WSA_WAIT_FAILED;
+ WSASetLastError (WSA_INVALID_PARAMETER);
+ }
if (ignore_count == 0 && wait_ret == WSA_WAIT_TIMEOUT) {
res = 0;
}
}
- g_mutex_lock (set->lock);
-
- gst_poll_check_ctrl_commands (set, res, &restarting);
-
- /* update the controllable state if needed */
- set->controllable = set->new_controllable;
-
- if (set->flushing) {
- /* we got woken up and we are flushing, we need to stop */
- errno = EBUSY;
- res = -1;
- break;
+ if (!is_timer) {
+ /* Applications needs to clear the control socket themselves for timer
+ * polls.
+ * For other polls, we need to clear the control socket. If there was only
+ * one socket with activity and it was the control socket, we need to
+ * restart */
+ if (release_all_wakeup (set) > 0 && res == 1)
+ restarting = TRUE;
}
- } while (restarting);
- set->waiting = FALSE;
+ /* we got woken up and we are flushing, we need to stop */
+ if (G_UNLIKELY (IS_FLUSHING (set)))
+ goto flushing;
+
+ } while (G_UNLIKELY (restarting));
- g_mutex_unlock (set->lock);
+ DEC_WAITING (set);
return res;
/* ERRORS */
already_waiting:
{
- g_mutex_unlock (set->lock);
+ GST_LOG ("%p: we are already waiting", set);
+ DEC_WAITING (set);
errno = EPERM;
return -1;
}
flushing:
{
- g_mutex_unlock (set->lock);
+ GST_LOG ("%p: we are flushing", set);
+ DEC_WAITING (set);
errno = EBUSY;
return -1;
}
#ifdef G_OS_WIN32
winsock_error:
{
- g_mutex_unlock (set->lock);
+ GST_LOG ("%p: winsock error", set);
+ g_mutex_unlock (&set->lock);
+ DEC_WAITING (set);
return -1;
}
#endif
* gst_poll_wait() will be affected by gst_poll_restart() and
* gst_poll_set_flushing().
*
- * Returns: %TRUE if the controllability of @set could be updated.
+ * This function only works for non-timer #GstPoll objects created with
+ * gst_poll_new().
*
- * Since: 0.10.18
+ * Returns: %TRUE if the controllability of @set could be updated.
*/
gboolean
gst_poll_set_controllable (GstPoll * set, gboolean controllable)
{
g_return_val_if_fail (set != NULL, FALSE);
+ g_return_val_if_fail (!set->timer, FALSE);
- g_mutex_lock (set->lock);
-
-#ifndef G_OS_WIN32
- if (controllable && set->control_read_fd.fd < 0) {
- gint control_sock[2];
+ GST_LOG ("%p: controllable : %d", set, controllable);
- if (socketpair (PF_UNIX, SOCK_STREAM, 0, control_sock) < 0)
- goto no_socket_pair;
-
- fcntl (control_sock[0], F_SETFL, O_NONBLOCK);
- fcntl (control_sock[1], F_SETFL, O_NONBLOCK);
-
- set->control_read_fd.fd = control_sock[0];
- set->control_write_fd.fd = control_sock[1];
-
- gst_poll_add_fd_unlocked (set, &set->control_read_fd);
- }
-
- if (set->control_read_fd.fd >= 0)
- gst_poll_fd_ctl_read_unlocked (set, &set->control_read_fd, controllable);
-#endif
-
- /* delay the change of the controllable state if we are waiting */
- set->new_controllable = controllable;
- if (!set->waiting)
- set->controllable = controllable;
-
- g_mutex_unlock (set->lock);
+ set->controllable = controllable;
return TRUE;
-
- /* ERRORS */
-#ifndef G_OS_WIN32
-no_socket_pair:
- {
- g_mutex_unlock (set->lock);
- return FALSE;
- }
-#endif
}
/**
*
* If @set is not controllable, then this call will have no effect.
*
- * Since: 0.10.18
+ * This function only works for non-timer #GstPoll objects created with
+ * gst_poll_new().
*/
void
gst_poll_restart (GstPoll * set)
{
g_return_if_fail (set != NULL);
+ g_return_if_fail (!set->timer);
- g_mutex_lock (set->lock);
-
- if (set->controllable && set->waiting) {
-#ifndef G_OS_WIN32
- /* if we are waiting, we can send the command, else we do not have to
- * bother, future calls will automatically pick up the new fdset */
- SEND_COMMAND (set, GST_POLL_CMD_WAKEUP);
-#else
- SetEvent (set->wakeup_event);
-#endif
+ if (set->controllable && GET_WAITING (set) > 0) {
+ /* we are controllable and waiting, wake up the waiter. The socket will be
+ * cleared by the _wait() thread and the poll will be restarted */
+ raise_wakeup (set);
}
-
- g_mutex_unlock (set->lock);
}
/**
*
* Unsetting the flushing state will restore normal operation of @set.
*
- * Since: 0.10.18
+ * This function only works for non-timer #GstPoll objects created with
+ * gst_poll_new().
*/
void
gst_poll_set_flushing (GstPoll * set, gboolean flushing)
{
g_return_if_fail (set != NULL);
+ g_return_if_fail (!set->timer);
- g_mutex_lock (set->lock);
+ GST_LOG ("%p: flushing: %d", set, flushing);
/* update the new state first */
- set->flushing = flushing;
+ SET_FLUSHING (set, flushing);
- if (flushing && set->controllable && set->waiting) {
+ if (flushing && set->controllable && GET_WAITING (set) > 0) {
/* we are flushing, controllable and waiting, wake up the waiter. When we
* stop the flushing operation we don't clear the wakeup fd here, this will
* happen in the _wait() thread. */
-#ifndef G_OS_WIN32
- SEND_COMMAND (set, GST_POLL_CMD_WAKEUP);
-#else
- SetEvent (set->wakeup_event);
-#endif
+ raise_wakeup (set);
}
+}
- g_mutex_unlock (set->lock);
+/**
+ * gst_poll_write_control:
+ * @set: a #GstPoll.
+ *
+ * Write a byte to the control socket of the controllable @set.
+ * This function is mostly useful for timer #GstPoll objects created with
+ * gst_poll_new_timer().
+ *
+ * It will make any current and future gst_poll_wait() function return with
+ * 1, meaning the control socket is set. After an equal amount of calls to
+ * gst_poll_read_control() have been performed, calls to gst_poll_wait() will
+ * block again until their timeout expired.
+ *
+ * This function only works for timer #GstPoll objects created with
+ * gst_poll_new_timer().
+ *
+ * Returns: %TRUE on success. %FALSE when when the byte could not be written.
+ * errno contains the detailed error code but will never be EAGAIN, EINTR or
+ * EWOULDBLOCK. %FALSE always signals a critical error.
+ */
+gboolean
+gst_poll_write_control (GstPoll * set)
+{
+ gboolean res;
+
+ g_return_val_if_fail (set != NULL, FALSE);
+ g_return_val_if_fail (set->timer, FALSE);
+
+ res = raise_wakeup (set);
+
+ return res;
+}
+
+/**
+ * gst_poll_read_control:
+ * @set: a #GstPoll.
+ *
+ * Read a byte from the control socket of the controllable @set.
+ *
+ * This function only works for timer #GstPoll objects created with
+ * gst_poll_new_timer().
+ *
+ * Returns: %TRUE on success. %FALSE when when there was no byte to read or
+ * reading the byte failed. If there was no byte to read, and only then, errno
+ * will contain EWOULDBLOCK or EAGAIN. For all other values of errno this always signals a
+ * critical error.
+ */
+gboolean
+gst_poll_read_control (GstPoll * set)
+{
+ gboolean res;
+
+ g_return_val_if_fail (set != NULL, FALSE);
+ g_return_val_if_fail (set->timer, FALSE);
+
+ res = release_wakeup (set);
+
+ return res;
}