gstreamer/gst/gstscheduler.c

1043 lines
27 KiB
C
Raw Normal View History

/* GStreamer
* Copyright (C) 1999,2000 Erik Walthinsen <omega@cse.ogi.edu>
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
* 2000 Wim Taymans <wim.taymans@chello.be>
*
* gstscheduler.c: Default scheduling code for most cases
*
* This library is free software; you can redistribute it and/or
* modify it under the terms of the GNU Library General Public
* License as published by the Free Software Foundation; either
* version 2 of the License, or (at your option) any later version.
*
* This library is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Library General Public License for more details.
*
* 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.
*/
#include "gst_private.h"
#include "gstsystemclock.h"
#include "gstscheduler.h"
#include "gstinfo.h"
#include "gstregistrypool.h"
static void gst_scheduler_class_init (GstSchedulerClass * klass);
static void gst_scheduler_init (GstScheduler * sched);
static void gst_scheduler_dispose (GObject * object);
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
static GstObjectClass *parent_class = NULL;
static gchar *_default_name = NULL;
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
GType
gst_scheduler_get_type (void)
{
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
static GType _gst_scheduler_type = 0;
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
if (!_gst_scheduler_type) {
static const GTypeInfo scheduler_info = {
sizeof (GstSchedulerClass),
NULL,
NULL,
(GClassInitFunc) gst_scheduler_class_init,
NULL,
NULL,
sizeof (GstScheduler),
0,
(GInstanceInitFunc) gst_scheduler_init,
NULL
};
_gst_scheduler_type =
g_type_register_static (GST_TYPE_OBJECT, "GstScheduler",
&scheduler_info, G_TYPE_FLAG_ABSTRACT);
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
}
return _gst_scheduler_type;
}
static void
gst_scheduler_class_init (GstSchedulerClass * klass)
{
GObjectClass *gobject_class;
gobject_class = (GObjectClass *) klass;
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
parent_class = g_type_class_ref (GST_TYPE_OBJECT);
gobject_class->dispose = GST_DEBUG_FUNCPTR (gst_scheduler_dispose);
}
static void
gst_scheduler_init (GstScheduler * sched)
{
sched->clock_providers = NULL;
sched->clock_receivers = NULL;
sched->schedulers = NULL;
sched->state = GST_SCHEDULER_STATE_NONE;
sched->parent = NULL;
sched->parent_sched = NULL;
sched->clock = NULL;
}
static void
gst_scheduler_dispose (GObject * object)
{
GstScheduler *sched = GST_SCHEDULER (object);
/* thse lists should all be NULL */
GST_DEBUG ("scheduler %p dispose %p %p %p",
object,
sched->clock_providers, sched->clock_receivers, sched->schedulers);
gst_object_replace ((GstObject **) & sched->current_clock, NULL);
gst_object_replace ((GstObject **) & sched->clock, NULL);
docs/pwg/advanced-types.xml: Fix description for buffer-frames=0. Original commit message from CVS: * docs/pwg/advanced-types.xml: Fix description for buffer-frames=0. * docs/gst/tmpl/gstbin.sgml: * gst/gstbin.c: (gst_bin_child_state_change_func), (gst_bin_change_state), (gst_bin_change_state_norecurse): * gst/gstbin.h: * testsuite/threads/Makefile.am: * testsuite/threads/threadi.c: (cb_timeout), (cb_quit), (cb_eos), (cb_state), (cb_play), (main): Fix non-recursive state changes to *really* change the state of the object, and not just call parent_class->state_change. Fix a lot of lockups caused by this. Fixes #132775. Add test for the problem. Also enable test to show #142588 (fixed). * gst/gstthread.c: (gst_thread_change_state), (gst_thread_child_state_change): Don't exit the thread if we go to NULL and are inside thread context. Instead, return control to the main thread context and exit from there. * gst/gstelement.c: (gst_element_disable_threadsafe_properties): Don't unset virtual functions, since those may still be used. That's not necessarily correct, but suffices for now. * configure.ac: * testsuite/Makefile.am: * testsuite/pad/Makefile.am: * testsuite/pad/chainnopull.c: (gst_test_sink_class_init), (gst_test_sink_base_init), (gst_test_sink_chain), (gst_test_sink_init), (main): * testsuite/pad/getnopush.c: (gst_test_src_class_init), (gst_test_src_base_init), (gst_test_src_get), (gst_test_src_init), (main): * testsuite/pad/link.c: (gst_test_element_class_init), (gst_test_element_base_init), (gst_test_src_get), (gst_test_src_loop), (gst_test_src_init), (gst_test_filter_chain), (gst_test_filter_loop), (gst_test_filter_init), (gst_test_sink_chain), (gst_test_sink_loop), (gst_test_sink_init), (cb_error), (main): Add tests to show #150546. Pass, but should fail (currently disabled from the testsuite). * gst/gstscheduler.c: (gst_scheduler_dispose): Dereference child schedulers on dispose (#94464). * testsuite/bytestream/filepadsink.c: (gst_fp_sink_init): Fix typo. * testsuite/threads/thread.c: (main): Add more debug.
2005-01-31 15:51:19 +00:00
/* kids are held reference to, so dereference here. */
while (sched->schedulers != NULL) {
gst_scheduler_remove_scheduler (sched,
GST_SCHEDULER (sched->schedulers->data));
}
G_OBJECT_CLASS (parent_class)->dispose (object);
}
/**
* gst_scheduler_setup:
* @sched: the scheduler
*
* Prepare the scheduler.
*/
void
gst_scheduler_setup (GstScheduler * sched)
{
GstSchedulerClass *sclass;
This is an attempt at not segfaulting on errors but reporting some usefull info instead. Original commit message from CVS: This is an attempt at not segfaulting on errors but reporting some usefull info instead. - bin changes so errors can propagate. - changed the _FAST macros to _CAST because that is what they do. - removed all references to cothreads out of the core, they are really a scheduler issue, handler with a sched_private gpointer. - added a live buffer count, for debugging buffer leaks. - added error checking in gst_scheduler_state_transition this solves the "out of cothreads" problem. - GST_ELEMENT_NO_ENTRY == GST_ELEMENT_INFINITE_LOOP - added 2 private element flasg for use by the scheduler (_COTHREAD_STOPPING) is now - added scheduler entry points: - _yield : to create possible scheduling points. - _interrupt: to stop execution of an element. - _error: to signal en error condition to the scheduler. - improved error messages for pads. - signal gst_element_error where appropriate. - added the a new bin to the parent before entering it so one can reference its children. - queue memleak fixes on dispose. - added possible deadlock detection in queue (turned off be default) - GstBasicScheduler is a real class of its own now, hiding its internal variables. - GST_ELEMENT_IS_COTHREAD_STOPPING is gone. either call explicit _yield operations, or make a sane loop. - Better state change handling in filesrc. Better error reporting/recovery too. - updated core plugins. - detect non decoupled elements on scheduler boundries and error.
2001-12-22 21:18:17 +00:00
g_return_if_fail (GST_IS_SCHEDULER (sched));
sclass = GST_SCHEDULER_GET_CLASS (sched);
if (sclass->setup)
sclass->setup (sched);
}
/**
* gst_scheduler_reset:
* @sched: a #GstScheduler to reset.
*
* Reset the schedulers.
*/
void
gst_scheduler_reset (GstScheduler * sched)
{
GstSchedulerClass *sclass;
This is an attempt at not segfaulting on errors but reporting some usefull info instead. Original commit message from CVS: This is an attempt at not segfaulting on errors but reporting some usefull info instead. - bin changes so errors can propagate. - changed the _FAST macros to _CAST because that is what they do. - removed all references to cothreads out of the core, they are really a scheduler issue, handler with a sched_private gpointer. - added a live buffer count, for debugging buffer leaks. - added error checking in gst_scheduler_state_transition this solves the "out of cothreads" problem. - GST_ELEMENT_NO_ENTRY == GST_ELEMENT_INFINITE_LOOP - added 2 private element flasg for use by the scheduler (_COTHREAD_STOPPING) is now - added scheduler entry points: - _yield : to create possible scheduling points. - _interrupt: to stop execution of an element. - _error: to signal en error condition to the scheduler. - improved error messages for pads. - signal gst_element_error where appropriate. - added the a new bin to the parent before entering it so one can reference its children. - queue memleak fixes on dispose. - added possible deadlock detection in queue (turned off be default) - GstBasicScheduler is a real class of its own now, hiding its internal variables. - GST_ELEMENT_IS_COTHREAD_STOPPING is gone. either call explicit _yield operations, or make a sane loop. - Better state change handling in filesrc. Better error reporting/recovery too. - updated core plugins. - detect non decoupled elements on scheduler boundries and error.
2001-12-22 21:18:17 +00:00
g_return_if_fail (GST_IS_SCHEDULER (sched));
sclass = GST_SCHEDULER_GET_CLASS (sched);
if (sclass->reset)
sclass->reset (sched);
}
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
/**
* gst_scheduler_pad_link:
* @sched: the scheduler
* @srcpad: the srcpad to link
* @sinkpad: the sinkpad to link to
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
*
* Links the srcpad to the given sinkpad.
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
*/
void
gst_scheduler_pad_link (GstScheduler * sched, GstPad * srcpad, GstPad * sinkpad)
{
GstSchedulerClass *sclass;
This is an attempt at not segfaulting on errors but reporting some usefull info instead. Original commit message from CVS: This is an attempt at not segfaulting on errors but reporting some usefull info instead. - bin changes so errors can propagate. - changed the _FAST macros to _CAST because that is what they do. - removed all references to cothreads out of the core, they are really a scheduler issue, handler with a sched_private gpointer. - added a live buffer count, for debugging buffer leaks. - added error checking in gst_scheduler_state_transition this solves the "out of cothreads" problem. - GST_ELEMENT_NO_ENTRY == GST_ELEMENT_INFINITE_LOOP - added 2 private element flasg for use by the scheduler (_COTHREAD_STOPPING) is now - added scheduler entry points: - _yield : to create possible scheduling points. - _interrupt: to stop execution of an element. - _error: to signal en error condition to the scheduler. - improved error messages for pads. - signal gst_element_error where appropriate. - added the a new bin to the parent before entering it so one can reference its children. - queue memleak fixes on dispose. - added possible deadlock detection in queue (turned off be default) - GstBasicScheduler is a real class of its own now, hiding its internal variables. - GST_ELEMENT_IS_COTHREAD_STOPPING is gone. either call explicit _yield operations, or make a sane loop. - Better state change handling in filesrc. Better error reporting/recovery too. - updated core plugins. - detect non decoupled elements on scheduler boundries and error.
2001-12-22 21:18:17 +00:00
g_return_if_fail (GST_IS_SCHEDULER (sched));
g_return_if_fail (GST_IS_PAD (srcpad));
g_return_if_fail (GST_IS_PAD (sinkpad));
sclass = GST_SCHEDULER_GET_CLASS (sched);
if (sclass->pad_link)
sclass->pad_link (sched, srcpad, sinkpad);
}
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
/**
* gst_scheduler_pad_unlink:
* @sched: the scheduler
* @srcpad: the srcpad to unlink
* @sinkpad: the sinkpad to unlink from
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
*
* Unlinks the srcpad from the given sinkpad.
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
*/
void
gst_scheduler_pad_unlink (GstScheduler * sched, GstPad * srcpad,
GstPad * sinkpad)
{
GstSchedulerClass *sclass;
This is an attempt at not segfaulting on errors but reporting some usefull info instead. Original commit message from CVS: This is an attempt at not segfaulting on errors but reporting some usefull info instead. - bin changes so errors can propagate. - changed the _FAST macros to _CAST because that is what they do. - removed all references to cothreads out of the core, they are really a scheduler issue, handler with a sched_private gpointer. - added a live buffer count, for debugging buffer leaks. - added error checking in gst_scheduler_state_transition this solves the "out of cothreads" problem. - GST_ELEMENT_NO_ENTRY == GST_ELEMENT_INFINITE_LOOP - added 2 private element flasg for use by the scheduler (_COTHREAD_STOPPING) is now - added scheduler entry points: - _yield : to create possible scheduling points. - _interrupt: to stop execution of an element. - _error: to signal en error condition to the scheduler. - improved error messages for pads. - signal gst_element_error where appropriate. - added the a new bin to the parent before entering it so one can reference its children. - queue memleak fixes on dispose. - added possible deadlock detection in queue (turned off be default) - GstBasicScheduler is a real class of its own now, hiding its internal variables. - GST_ELEMENT_IS_COTHREAD_STOPPING is gone. either call explicit _yield operations, or make a sane loop. - Better state change handling in filesrc. Better error reporting/recovery too. - updated core plugins. - detect non decoupled elements on scheduler boundries and error.
2001-12-22 21:18:17 +00:00
g_return_if_fail (GST_IS_SCHEDULER (sched));
g_return_if_fail (GST_IS_PAD (srcpad));
g_return_if_fail (GST_IS_PAD (sinkpad));
sclass = GST_SCHEDULER_GET_CLASS (sched);
if (sclass->pad_unlink)
sclass->pad_unlink (sched, srcpad, sinkpad);
}
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
/**
* gst_scheduler_pad_select:
* @sched: the scheduler
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
* @padlist: the padlist to select on
*
* register the given padlist for a select operation.
*
* Returns: the pad which received a buffer.
*/
GstPad *
gst_scheduler_pad_select (GstScheduler * sched, GList * padlist)
{
g_return_val_if_fail (GST_IS_SCHEDULER (sched), NULL);
g_return_val_if_fail (padlist != NULL, NULL);
This is an attempt at not segfaulting on errors but reporting some usefull info instead. Original commit message from CVS: This is an attempt at not segfaulting on errors but reporting some usefull info instead. - bin changes so errors can propagate. - changed the _FAST macros to _CAST because that is what they do. - removed all references to cothreads out of the core, they are really a scheduler issue, handler with a sched_private gpointer. - added a live buffer count, for debugging buffer leaks. - added error checking in gst_scheduler_state_transition this solves the "out of cothreads" problem. - GST_ELEMENT_NO_ENTRY == GST_ELEMENT_INFINITE_LOOP - added 2 private element flasg for use by the scheduler (_COTHREAD_STOPPING) is now - added scheduler entry points: - _yield : to create possible scheduling points. - _interrupt: to stop execution of an element. - _error: to signal en error condition to the scheduler. - improved error messages for pads. - signal gst_element_error where appropriate. - added the a new bin to the parent before entering it so one can reference its children. - queue memleak fixes on dispose. - added possible deadlock detection in queue (turned off be default) - GstBasicScheduler is a real class of its own now, hiding its internal variables. - GST_ELEMENT_IS_COTHREAD_STOPPING is gone. either call explicit _yield operations, or make a sane loop. - Better state change handling in filesrc. Better error reporting/recovery too. - updated core plugins. - detect non decoupled elements on scheduler boundries and error.
2001-12-22 21:18:17 +00:00
return NULL;
}
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
/**
* gst_scheduler_add_element:
* @sched: the scheduler
* @element: the element to add to the scheduler
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
*
* Add an element to the scheduler.
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
*/
void
gst_scheduler_add_element (GstScheduler * sched, GstElement * element)
{
GstSchedulerClass *sclass;
gboolean redistribute_clock = FALSE;
This is an attempt at not segfaulting on errors but reporting some usefull info instead. Original commit message from CVS: This is an attempt at not segfaulting on errors but reporting some usefull info instead. - bin changes so errors can propagate. - changed the _FAST macros to _CAST because that is what they do. - removed all references to cothreads out of the core, they are really a scheduler issue, handler with a sched_private gpointer. - added a live buffer count, for debugging buffer leaks. - added error checking in gst_scheduler_state_transition this solves the "out of cothreads" problem. - GST_ELEMENT_NO_ENTRY == GST_ELEMENT_INFINITE_LOOP - added 2 private element flasg for use by the scheduler (_COTHREAD_STOPPING) is now - added scheduler entry points: - _yield : to create possible scheduling points. - _interrupt: to stop execution of an element. - _error: to signal en error condition to the scheduler. - improved error messages for pads. - signal gst_element_error where appropriate. - added the a new bin to the parent before entering it so one can reference its children. - queue memleak fixes on dispose. - added possible deadlock detection in queue (turned off be default) - GstBasicScheduler is a real class of its own now, hiding its internal variables. - GST_ELEMENT_IS_COTHREAD_STOPPING is gone. either call explicit _yield operations, or make a sane loop. - Better state change handling in filesrc. Better error reporting/recovery too. - updated core plugins. - detect non decoupled elements on scheduler boundries and error.
2001-12-22 21:18:17 +00:00
g_return_if_fail (GST_IS_SCHEDULER (sched));
g_return_if_fail (GST_IS_ELEMENT (element));
/* if it's already in this scheduler, don't bother doing anything */
if (GST_ELEMENT_SCHED (element) == sched) {
GST_CAT_DEBUG (GST_CAT_SCHEDULING, "element %s already in scheduler %p",
GST_ELEMENT_NAME (element), sched);
return;
}
/* if it's not inside this scheduler, it has to be NULL */
g_assert (GST_ELEMENT_SCHED (element) == NULL);
if (gst_element_provides_clock (element)) {
sched->clock_providers = g_list_prepend (sched->clock_providers, element);
GST_CAT_DEBUG (GST_CAT_CLOCK, "added clock provider %s",
GST_ELEMENT_NAME (element));
redistribute_clock = TRUE;
}
if (gst_element_requires_clock (element)) {
sched->clock_receivers = g_list_prepend (sched->clock_receivers, element);
GST_CAT_DEBUG (GST_CAT_CLOCK, "added clock receiver %s",
GST_ELEMENT_NAME (element));
redistribute_clock = TRUE;
}
gst_element_set_scheduler (element, sched);
if (redistribute_clock) {
GstClock *clock;
clock = gst_scheduler_get_clock (sched);
gst_scheduler_set_clock (sched, clock);
}
sclass = GST_SCHEDULER_GET_CLASS (sched);
if (sclass->add_element)
sclass->add_element (sched, element);
}
/**
* gst_scheduler_remove_element:
* @sched: the scheduler
* @element: the element to remove
*
* Remove an element from the scheduler.
*/
void
gst_scheduler_remove_element (GstScheduler * sched, GstElement * element)
{
GstSchedulerClass *sclass;
GList *link;
gboolean redistribute_clock = FALSE;
g_return_if_fail (GST_IS_SCHEDULER (sched));
g_return_if_fail (GST_IS_ELEMENT (element));
link = g_list_find (sched->clock_providers, element);
if (link) {
sched->clock_providers = g_list_delete_link (sched->clock_providers, link);
GST_CAT_DEBUG (GST_CAT_CLOCK, "removed clock provider %s",
GST_ELEMENT_NAME (element));
redistribute_clock = TRUE;
}
link = g_list_find (sched->clock_receivers, element);
if (link) {
sched->clock_receivers = g_list_delete_link (sched->clock_receivers, link);
GST_CAT_DEBUG (GST_CAT_CLOCK, "removed clock receiver %s",
GST_ELEMENT_NAME (element));
redistribute_clock = TRUE;
}
if (redistribute_clock) {
GstClock *clock;
clock = gst_scheduler_get_clock (sched);
gst_scheduler_set_clock (sched, clock);
}
sclass = GST_SCHEDULER_GET_CLASS (sched);
if (sclass->remove_element)
sclass->remove_element (sched, element);
gst_element_set_scheduler (element, NULL);
}
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
/**
* gst_scheduler_state_transition:
* @sched: the scheduler
* @element: the element with the state transition
* @transition: the state transition
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
*
* Tell the scheduler that an element changed its state.
*
* Returns: a GstElementStateReturn indicating success or failure
* of the state transition.
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
*/
This is an attempt at not segfaulting on errors but reporting some usefull info instead. Original commit message from CVS: This is an attempt at not segfaulting on errors but reporting some usefull info instead. - bin changes so errors can propagate. - changed the _FAST macros to _CAST because that is what they do. - removed all references to cothreads out of the core, they are really a scheduler issue, handler with a sched_private gpointer. - added a live buffer count, for debugging buffer leaks. - added error checking in gst_scheduler_state_transition this solves the "out of cothreads" problem. - GST_ELEMENT_NO_ENTRY == GST_ELEMENT_INFINITE_LOOP - added 2 private element flasg for use by the scheduler (_COTHREAD_STOPPING) is now - added scheduler entry points: - _yield : to create possible scheduling points. - _interrupt: to stop execution of an element. - _error: to signal en error condition to the scheduler. - improved error messages for pads. - signal gst_element_error where appropriate. - added the a new bin to the parent before entering it so one can reference its children. - queue memleak fixes on dispose. - added possible deadlock detection in queue (turned off be default) - GstBasicScheduler is a real class of its own now, hiding its internal variables. - GST_ELEMENT_IS_COTHREAD_STOPPING is gone. either call explicit _yield operations, or make a sane loop. - Better state change handling in filesrc. Better error reporting/recovery too. - updated core plugins. - detect non decoupled elements on scheduler boundries and error.
2001-12-22 21:18:17 +00:00
GstElementStateReturn
gst_scheduler_state_transition (GstScheduler * sched, GstElement * element,
gint transition)
{
GstSchedulerClass *sclass;
This is an attempt at not segfaulting on errors but reporting some usefull info instead. Original commit message from CVS: This is an attempt at not segfaulting on errors but reporting some usefull info instead. - bin changes so errors can propagate. - changed the _FAST macros to _CAST because that is what they do. - removed all references to cothreads out of the core, they are really a scheduler issue, handler with a sched_private gpointer. - added a live buffer count, for debugging buffer leaks. - added error checking in gst_scheduler_state_transition this solves the "out of cothreads" problem. - GST_ELEMENT_NO_ENTRY == GST_ELEMENT_INFINITE_LOOP - added 2 private element flasg for use by the scheduler (_COTHREAD_STOPPING) is now - added scheduler entry points: - _yield : to create possible scheduling points. - _interrupt: to stop execution of an element. - _error: to signal en error condition to the scheduler. - improved error messages for pads. - signal gst_element_error where appropriate. - added the a new bin to the parent before entering it so one can reference its children. - queue memleak fixes on dispose. - added possible deadlock detection in queue (turned off be default) - GstBasicScheduler is a real class of its own now, hiding its internal variables. - GST_ELEMENT_IS_COTHREAD_STOPPING is gone. either call explicit _yield operations, or make a sane loop. - Better state change handling in filesrc. Better error reporting/recovery too. - updated core plugins. - detect non decoupled elements on scheduler boundries and error.
2001-12-22 21:18:17 +00:00
g_return_val_if_fail (GST_IS_SCHEDULER (sched), GST_STATE_FAILURE);
g_return_val_if_fail (GST_IS_ELEMENT (element), GST_STATE_FAILURE);
if (element == sched->parent && sched->parent_sched == NULL) {
/* FIXME is distributing the clock in the state change still needed
* when we distribute as soon as we add/remove elements? I think not.*/
switch (transition) {
case GST_STATE_READY_TO_PAUSED:
{
GstClock *clock = gst_scheduler_get_clock (sched);
GST_CAT_DEBUG (GST_CAT_CLOCK,
"scheduler READY to PAUSED clock is %p (%s)", clock,
(clock ? GST_OBJECT_NAME (clock) : "nil"));
gst_scheduler_set_clock (sched, clock);
break;
}
}
}
sclass = GST_SCHEDULER_GET_CLASS (sched);
if (sclass->state_transition)
return sclass->state_transition (sched, element, transition);
This is an attempt at not segfaulting on errors but reporting some usefull info instead. Original commit message from CVS: This is an attempt at not segfaulting on errors but reporting some usefull info instead. - bin changes so errors can propagate. - changed the _FAST macros to _CAST because that is what they do. - removed all references to cothreads out of the core, they are really a scheduler issue, handler with a sched_private gpointer. - added a live buffer count, for debugging buffer leaks. - added error checking in gst_scheduler_state_transition this solves the "out of cothreads" problem. - GST_ELEMENT_NO_ENTRY == GST_ELEMENT_INFINITE_LOOP - added 2 private element flasg for use by the scheduler (_COTHREAD_STOPPING) is now - added scheduler entry points: - _yield : to create possible scheduling points. - _interrupt: to stop execution of an element. - _error: to signal en error condition to the scheduler. - improved error messages for pads. - signal gst_element_error where appropriate. - added the a new bin to the parent before entering it so one can reference its children. - queue memleak fixes on dispose. - added possible deadlock detection in queue (turned off be default) - GstBasicScheduler is a real class of its own now, hiding its internal variables. - GST_ELEMENT_IS_COTHREAD_STOPPING is gone. either call explicit _yield operations, or make a sane loop. - Better state change handling in filesrc. Better error reporting/recovery too. - updated core plugins. - detect non decoupled elements on scheduler boundries and error.
2001-12-22 21:18:17 +00:00
return GST_STATE_SUCCESS;
}
/**
* gst_scheduler_scheduling_change:
* @sched: the scheduler
* @element: the element that changed its scheduling strategy
*
* Tell the scheduler that an element changed its scheduling strategy.
* An element could, for example, change its loop function or changes
* from a loop based element to a chain based element.
*/
void
gst_scheduler_scheduling_change (GstScheduler * sched, GstElement * element)
{
GstSchedulerClass *sclass;
g_return_if_fail (GST_IS_SCHEDULER (sched));
g_return_if_fail (GST_IS_ELEMENT (element));
sclass = GST_SCHEDULER_GET_CLASS (sched);
if (sclass->scheduling_change)
sclass->scheduling_change (sched, element);
}
/**
* gst_scheduler_add_scheduler:
* @sched: a #GstScheduler to add to
* @sched2: the #GstScheduler to add
*
* Notifies the scheduler that it has to monitor this scheduler.
*/
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
void
gst_scheduler_add_scheduler (GstScheduler * sched, GstScheduler * sched2)
{
GstSchedulerClass *sclass;
This is an attempt at not segfaulting on errors but reporting some usefull info instead. Original commit message from CVS: This is an attempt at not segfaulting on errors but reporting some usefull info instead. - bin changes so errors can propagate. - changed the _FAST macros to _CAST because that is what they do. - removed all references to cothreads out of the core, they are really a scheduler issue, handler with a sched_private gpointer. - added a live buffer count, for debugging buffer leaks. - added error checking in gst_scheduler_state_transition this solves the "out of cothreads" problem. - GST_ELEMENT_NO_ENTRY == GST_ELEMENT_INFINITE_LOOP - added 2 private element flasg for use by the scheduler (_COTHREAD_STOPPING) is now - added scheduler entry points: - _yield : to create possible scheduling points. - _interrupt: to stop execution of an element. - _error: to signal en error condition to the scheduler. - improved error messages for pads. - signal gst_element_error where appropriate. - added the a new bin to the parent before entering it so one can reference its children. - queue memleak fixes on dispose. - added possible deadlock detection in queue (turned off be default) - GstBasicScheduler is a real class of its own now, hiding its internal variables. - GST_ELEMENT_IS_COTHREAD_STOPPING is gone. either call explicit _yield operations, or make a sane loop. - Better state change handling in filesrc. Better error reporting/recovery too. - updated core plugins. - detect non decoupled elements on scheduler boundries and error.
2001-12-22 21:18:17 +00:00
g_return_if_fail (GST_IS_SCHEDULER (sched));
g_return_if_fail (GST_IS_SCHEDULER (sched2));
g_return_if_fail (sched2->parent_sched == NULL);
GST_DEBUG ("gstscheduler: %p add scheduler %p", sched, sched2);
gst_object_ref (GST_OBJECT (sched2));
gst_object_ref (GST_OBJECT (sched));
This is an attempt at not segfaulting on errors but reporting some usefull info instead. Original commit message from CVS: This is an attempt at not segfaulting on errors but reporting some usefull info instead. - bin changes so errors can propagate. - changed the _FAST macros to _CAST because that is what they do. - removed all references to cothreads out of the core, they are really a scheduler issue, handler with a sched_private gpointer. - added a live buffer count, for debugging buffer leaks. - added error checking in gst_scheduler_state_transition this solves the "out of cothreads" problem. - GST_ELEMENT_NO_ENTRY == GST_ELEMENT_INFINITE_LOOP - added 2 private element flasg for use by the scheduler (_COTHREAD_STOPPING) is now - added scheduler entry points: - _yield : to create possible scheduling points. - _interrupt: to stop execution of an element. - _error: to signal en error condition to the scheduler. - improved error messages for pads. - signal gst_element_error where appropriate. - added the a new bin to the parent before entering it so one can reference its children. - queue memleak fixes on dispose. - added possible deadlock detection in queue (turned off be default) - GstBasicScheduler is a real class of its own now, hiding its internal variables. - GST_ELEMENT_IS_COTHREAD_STOPPING is gone. either call explicit _yield operations, or make a sane loop. - Better state change handling in filesrc. Better error reporting/recovery too. - updated core plugins. - detect non decoupled elements on scheduler boundries and error.
2001-12-22 21:18:17 +00:00
sched->schedulers = g_list_prepend (sched->schedulers, sched2);
sched2->parent_sched = sched;
sclass = GST_SCHEDULER_GET_CLASS (sched);
if (sclass->add_scheduler)
sclass->add_scheduler (sched, sched2);
}
/**
* gst_scheduler_remove_scheduler:
* @sched: the scheduler
* @sched2: the scheduler to remove
*
a Notifies the scheduler that it can stop monitoring this scheduler.
*/
void
gst_scheduler_remove_scheduler (GstScheduler * sched, GstScheduler * sched2)
{
GstSchedulerClass *sclass;
g_return_if_fail (GST_IS_SCHEDULER (sched));
g_return_if_fail (GST_IS_SCHEDULER (sched2));
g_return_if_fail (sched2->parent_sched == sched);
GST_DEBUG ("gstscheduler: %p remove scheduler %p", sched, sched2);
sclass = GST_SCHEDULER_GET_CLASS (sched);
if (sclass->remove_scheduler)
sclass->remove_scheduler (sched, sched2);
sched->schedulers = g_list_remove (sched->schedulers, sched2);
sched2->parent_sched = NULL;
gst_object_unref (GST_OBJECT (sched2));
gst_object_unref (GST_OBJECT (sched));
}
/**
* gst_scheduler_lock_element:
* @sched: the scheduler
* @element: the element to lock
*
* Acquire a lock on the given element in the given scheduler.
*/
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
void
gst_scheduler_lock_element (GstScheduler * sched, GstElement * element)
{
This is an attempt at not segfaulting on errors but reporting some usefull info instead. Original commit message from CVS: This is an attempt at not segfaulting on errors but reporting some usefull info instead. - bin changes so errors can propagate. - changed the _FAST macros to _CAST because that is what they do. - removed all references to cothreads out of the core, they are really a scheduler issue, handler with a sched_private gpointer. - added a live buffer count, for debugging buffer leaks. - added error checking in gst_scheduler_state_transition this solves the "out of cothreads" problem. - GST_ELEMENT_NO_ENTRY == GST_ELEMENT_INFINITE_LOOP - added 2 private element flasg for use by the scheduler (_COTHREAD_STOPPING) is now - added scheduler entry points: - _yield : to create possible scheduling points. - _interrupt: to stop execution of an element. - _error: to signal en error condition to the scheduler. - improved error messages for pads. - signal gst_element_error where appropriate. - added the a new bin to the parent before entering it so one can reference its children. - queue memleak fixes on dispose. - added possible deadlock detection in queue (turned off be default) - GstBasicScheduler is a real class of its own now, hiding its internal variables. - GST_ELEMENT_IS_COTHREAD_STOPPING is gone. either call explicit _yield operations, or make a sane loop. - Better state change handling in filesrc. Better error reporting/recovery too. - updated core plugins. - detect non decoupled elements on scheduler boundries and error.
2001-12-22 21:18:17 +00:00
g_return_if_fail (GST_IS_SCHEDULER (sched));
g_return_if_fail (GST_IS_ELEMENT (element));
}
/**
* gst_scheduler_unlock_element:
* @sched: the scheduler
* @element: the element to unlock
*
* Release the lock on the given element in the given scheduler.
*/
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
void
gst_scheduler_unlock_element (GstScheduler * sched, GstElement * element)
{
GstSchedulerClass *sclass;
This is an attempt at not segfaulting on errors but reporting some usefull info instead. Original commit message from CVS: This is an attempt at not segfaulting on errors but reporting some usefull info instead. - bin changes so errors can propagate. - changed the _FAST macros to _CAST because that is what they do. - removed all references to cothreads out of the core, they are really a scheduler issue, handler with a sched_private gpointer. - added a live buffer count, for debugging buffer leaks. - added error checking in gst_scheduler_state_transition this solves the "out of cothreads" problem. - GST_ELEMENT_NO_ENTRY == GST_ELEMENT_INFINITE_LOOP - added 2 private element flasg for use by the scheduler (_COTHREAD_STOPPING) is now - added scheduler entry points: - _yield : to create possible scheduling points. - _interrupt: to stop execution of an element. - _error: to signal en error condition to the scheduler. - improved error messages for pads. - signal gst_element_error where appropriate. - added the a new bin to the parent before entering it so one can reference its children. - queue memleak fixes on dispose. - added possible deadlock detection in queue (turned off be default) - GstBasicScheduler is a real class of its own now, hiding its internal variables. - GST_ELEMENT_IS_COTHREAD_STOPPING is gone. either call explicit _yield operations, or make a sane loop. - Better state change handling in filesrc. Better error reporting/recovery too. - updated core plugins. - detect non decoupled elements on scheduler boundries and error.
2001-12-22 21:18:17 +00:00
g_return_if_fail (GST_IS_SCHEDULER (sched));
g_return_if_fail (GST_IS_ELEMENT (element));
sclass = GST_SCHEDULER_GET_CLASS (sched);
}
This is an attempt at not segfaulting on errors but reporting some usefull info instead. Original commit message from CVS: This is an attempt at not segfaulting on errors but reporting some usefull info instead. - bin changes so errors can propagate. - changed the _FAST macros to _CAST because that is what they do. - removed all references to cothreads out of the core, they are really a scheduler issue, handler with a sched_private gpointer. - added a live buffer count, for debugging buffer leaks. - added error checking in gst_scheduler_state_transition this solves the "out of cothreads" problem. - GST_ELEMENT_NO_ENTRY == GST_ELEMENT_INFINITE_LOOP - added 2 private element flasg for use by the scheduler (_COTHREAD_STOPPING) is now - added scheduler entry points: - _yield : to create possible scheduling points. - _interrupt: to stop execution of an element. - _error: to signal en error condition to the scheduler. - improved error messages for pads. - signal gst_element_error where appropriate. - added the a new bin to the parent before entering it so one can reference its children. - queue memleak fixes on dispose. - added possible deadlock detection in queue (turned off be default) - GstBasicScheduler is a real class of its own now, hiding its internal variables. - GST_ELEMENT_IS_COTHREAD_STOPPING is gone. either call explicit _yield operations, or make a sane loop. - Better state change handling in filesrc. Better error reporting/recovery too. - updated core plugins. - detect non decoupled elements on scheduler boundries and error.
2001-12-22 21:18:17 +00:00
/**
* gst_scheduler_error:
* @sched: the scheduler
This is an attempt at not segfaulting on errors but reporting some usefull info instead. Original commit message from CVS: This is an attempt at not segfaulting on errors but reporting some usefull info instead. - bin changes so errors can propagate. - changed the _FAST macros to _CAST because that is what they do. - removed all references to cothreads out of the core, they are really a scheduler issue, handler with a sched_private gpointer. - added a live buffer count, for debugging buffer leaks. - added error checking in gst_scheduler_state_transition this solves the "out of cothreads" problem. - GST_ELEMENT_NO_ENTRY == GST_ELEMENT_INFINITE_LOOP - added 2 private element flasg for use by the scheduler (_COTHREAD_STOPPING) is now - added scheduler entry points: - _yield : to create possible scheduling points. - _interrupt: to stop execution of an element. - _error: to signal en error condition to the scheduler. - improved error messages for pads. - signal gst_element_error where appropriate. - added the a new bin to the parent before entering it so one can reference its children. - queue memleak fixes on dispose. - added possible deadlock detection in queue (turned off be default) - GstBasicScheduler is a real class of its own now, hiding its internal variables. - GST_ELEMENT_IS_COTHREAD_STOPPING is gone. either call explicit _yield operations, or make a sane loop. - Better state change handling in filesrc. Better error reporting/recovery too. - updated core plugins. - detect non decoupled elements on scheduler boundries and error.
2001-12-22 21:18:17 +00:00
* @element: the element with the error
*
* Tell the scheduler an element was in error
*/
void
gst_scheduler_error (GstScheduler * sched, GstElement * element)
This is an attempt at not segfaulting on errors but reporting some usefull info instead. Original commit message from CVS: This is an attempt at not segfaulting on errors but reporting some usefull info instead. - bin changes so errors can propagate. - changed the _FAST macros to _CAST because that is what they do. - removed all references to cothreads out of the core, they are really a scheduler issue, handler with a sched_private gpointer. - added a live buffer count, for debugging buffer leaks. - added error checking in gst_scheduler_state_transition this solves the "out of cothreads" problem. - GST_ELEMENT_NO_ENTRY == GST_ELEMENT_INFINITE_LOOP - added 2 private element flasg for use by the scheduler (_COTHREAD_STOPPING) is now - added scheduler entry points: - _yield : to create possible scheduling points. - _interrupt: to stop execution of an element. - _error: to signal en error condition to the scheduler. - improved error messages for pads. - signal gst_element_error where appropriate. - added the a new bin to the parent before entering it so one can reference its children. - queue memleak fixes on dispose. - added possible deadlock detection in queue (turned off be default) - GstBasicScheduler is a real class of its own now, hiding its internal variables. - GST_ELEMENT_IS_COTHREAD_STOPPING is gone. either call explicit _yield operations, or make a sane loop. - Better state change handling in filesrc. Better error reporting/recovery too. - updated core plugins. - detect non decoupled elements on scheduler boundries and error.
2001-12-22 21:18:17 +00:00
{
GstSchedulerClass *sclass;
This is an attempt at not segfaulting on errors but reporting some usefull info instead. Original commit message from CVS: This is an attempt at not segfaulting on errors but reporting some usefull info instead. - bin changes so errors can propagate. - changed the _FAST macros to _CAST because that is what they do. - removed all references to cothreads out of the core, they are really a scheduler issue, handler with a sched_private gpointer. - added a live buffer count, for debugging buffer leaks. - added error checking in gst_scheduler_state_transition this solves the "out of cothreads" problem. - GST_ELEMENT_NO_ENTRY == GST_ELEMENT_INFINITE_LOOP - added 2 private element flasg for use by the scheduler (_COTHREAD_STOPPING) is now - added scheduler entry points: - _yield : to create possible scheduling points. - _interrupt: to stop execution of an element. - _error: to signal en error condition to the scheduler. - improved error messages for pads. - signal gst_element_error where appropriate. - added the a new bin to the parent before entering it so one can reference its children. - queue memleak fixes on dispose. - added possible deadlock detection in queue (turned off be default) - GstBasicScheduler is a real class of its own now, hiding its internal variables. - GST_ELEMENT_IS_COTHREAD_STOPPING is gone. either call explicit _yield operations, or make a sane loop. - Better state change handling in filesrc. Better error reporting/recovery too. - updated core plugins. - detect non decoupled elements on scheduler boundries and error.
2001-12-22 21:18:17 +00:00
g_return_if_fail (GST_IS_SCHEDULER (sched));
g_return_if_fail (GST_IS_ELEMENT (element));
sclass = GST_SCHEDULER_GET_CLASS (sched);
if (sclass->error)
sclass->error (sched, element);
This is an attempt at not segfaulting on errors but reporting some usefull info instead. Original commit message from CVS: This is an attempt at not segfaulting on errors but reporting some usefull info instead. - bin changes so errors can propagate. - changed the _FAST macros to _CAST because that is what they do. - removed all references to cothreads out of the core, they are really a scheduler issue, handler with a sched_private gpointer. - added a live buffer count, for debugging buffer leaks. - added error checking in gst_scheduler_state_transition this solves the "out of cothreads" problem. - GST_ELEMENT_NO_ENTRY == GST_ELEMENT_INFINITE_LOOP - added 2 private element flasg for use by the scheduler (_COTHREAD_STOPPING) is now - added scheduler entry points: - _yield : to create possible scheduling points. - _interrupt: to stop execution of an element. - _error: to signal en error condition to the scheduler. - improved error messages for pads. - signal gst_element_error where appropriate. - added the a new bin to the parent before entering it so one can reference its children. - queue memleak fixes on dispose. - added possible deadlock detection in queue (turned off be default) - GstBasicScheduler is a real class of its own now, hiding its internal variables. - GST_ELEMENT_IS_COTHREAD_STOPPING is gone. either call explicit _yield operations, or make a sane loop. - Better state change handling in filesrc. Better error reporting/recovery too. - updated core plugins. - detect non decoupled elements on scheduler boundries and error.
2001-12-22 21:18:17 +00:00
}
/**
* gst_scheduler_yield:
* @sched: the scheduler
This is an attempt at not segfaulting on errors but reporting some usefull info instead. Original commit message from CVS: This is an attempt at not segfaulting on errors but reporting some usefull info instead. - bin changes so errors can propagate. - changed the _FAST macros to _CAST because that is what they do. - removed all references to cothreads out of the core, they are really a scheduler issue, handler with a sched_private gpointer. - added a live buffer count, for debugging buffer leaks. - added error checking in gst_scheduler_state_transition this solves the "out of cothreads" problem. - GST_ELEMENT_NO_ENTRY == GST_ELEMENT_INFINITE_LOOP - added 2 private element flasg for use by the scheduler (_COTHREAD_STOPPING) is now - added scheduler entry points: - _yield : to create possible scheduling points. - _interrupt: to stop execution of an element. - _error: to signal en error condition to the scheduler. - improved error messages for pads. - signal gst_element_error where appropriate. - added the a new bin to the parent before entering it so one can reference its children. - queue memleak fixes on dispose. - added possible deadlock detection in queue (turned off be default) - GstBasicScheduler is a real class of its own now, hiding its internal variables. - GST_ELEMENT_IS_COTHREAD_STOPPING is gone. either call explicit _yield operations, or make a sane loop. - Better state change handling in filesrc. Better error reporting/recovery too. - updated core plugins. - detect non decoupled elements on scheduler boundries and error.
2001-12-22 21:18:17 +00:00
* @element: the element requesting a yield
*
* Tell the scheduler to schedule another element.
*
* Returns: TRUE if the element should save its state, FALSE
* if the scheduler can perform this action itself.
This is an attempt at not segfaulting on errors but reporting some usefull info instead. Original commit message from CVS: This is an attempt at not segfaulting on errors but reporting some usefull info instead. - bin changes so errors can propagate. - changed the _FAST macros to _CAST because that is what they do. - removed all references to cothreads out of the core, they are really a scheduler issue, handler with a sched_private gpointer. - added a live buffer count, for debugging buffer leaks. - added error checking in gst_scheduler_state_transition this solves the "out of cothreads" problem. - GST_ELEMENT_NO_ENTRY == GST_ELEMENT_INFINITE_LOOP - added 2 private element flasg for use by the scheduler (_COTHREAD_STOPPING) is now - added scheduler entry points: - _yield : to create possible scheduling points. - _interrupt: to stop execution of an element. - _error: to signal en error condition to the scheduler. - improved error messages for pads. - signal gst_element_error where appropriate. - added the a new bin to the parent before entering it so one can reference its children. - queue memleak fixes on dispose. - added possible deadlock detection in queue (turned off be default) - GstBasicScheduler is a real class of its own now, hiding its internal variables. - GST_ELEMENT_IS_COTHREAD_STOPPING is gone. either call explicit _yield operations, or make a sane loop. - Better state change handling in filesrc. Better error reporting/recovery too. - updated core plugins. - detect non decoupled elements on scheduler boundries and error.
2001-12-22 21:18:17 +00:00
*/
gboolean
gst_scheduler_yield (GstScheduler * sched, GstElement * element)
This is an attempt at not segfaulting on errors but reporting some usefull info instead. Original commit message from CVS: This is an attempt at not segfaulting on errors but reporting some usefull info instead. - bin changes so errors can propagate. - changed the _FAST macros to _CAST because that is what they do. - removed all references to cothreads out of the core, they are really a scheduler issue, handler with a sched_private gpointer. - added a live buffer count, for debugging buffer leaks. - added error checking in gst_scheduler_state_transition this solves the "out of cothreads" problem. - GST_ELEMENT_NO_ENTRY == GST_ELEMENT_INFINITE_LOOP - added 2 private element flasg for use by the scheduler (_COTHREAD_STOPPING) is now - added scheduler entry points: - _yield : to create possible scheduling points. - _interrupt: to stop execution of an element. - _error: to signal en error condition to the scheduler. - improved error messages for pads. - signal gst_element_error where appropriate. - added the a new bin to the parent before entering it so one can reference its children. - queue memleak fixes on dispose. - added possible deadlock detection in queue (turned off be default) - GstBasicScheduler is a real class of its own now, hiding its internal variables. - GST_ELEMENT_IS_COTHREAD_STOPPING is gone. either call explicit _yield operations, or make a sane loop. - Better state change handling in filesrc. Better error reporting/recovery too. - updated core plugins. - detect non decoupled elements on scheduler boundries and error.
2001-12-22 21:18:17 +00:00
{
GstSchedulerClass *sclass;
g_return_val_if_fail (GST_IS_SCHEDULER (sched), TRUE);
g_return_val_if_fail (GST_IS_ELEMENT (element), TRUE);
This is an attempt at not segfaulting on errors but reporting some usefull info instead. Original commit message from CVS: This is an attempt at not segfaulting on errors but reporting some usefull info instead. - bin changes so errors can propagate. - changed the _FAST macros to _CAST because that is what they do. - removed all references to cothreads out of the core, they are really a scheduler issue, handler with a sched_private gpointer. - added a live buffer count, for debugging buffer leaks. - added error checking in gst_scheduler_state_transition this solves the "out of cothreads" problem. - GST_ELEMENT_NO_ENTRY == GST_ELEMENT_INFINITE_LOOP - added 2 private element flasg for use by the scheduler (_COTHREAD_STOPPING) is now - added scheduler entry points: - _yield : to create possible scheduling points. - _interrupt: to stop execution of an element. - _error: to signal en error condition to the scheduler. - improved error messages for pads. - signal gst_element_error where appropriate. - added the a new bin to the parent before entering it so one can reference its children. - queue memleak fixes on dispose. - added possible deadlock detection in queue (turned off be default) - GstBasicScheduler is a real class of its own now, hiding its internal variables. - GST_ELEMENT_IS_COTHREAD_STOPPING is gone. either call explicit _yield operations, or make a sane loop. - Better state change handling in filesrc. Better error reporting/recovery too. - updated core plugins. - detect non decoupled elements on scheduler boundries and error.
2001-12-22 21:18:17 +00:00
sclass = GST_SCHEDULER_GET_CLASS (sched);
if (sclass->yield)
return sclass->yield (sched, element);
return TRUE;
This is an attempt at not segfaulting on errors but reporting some usefull info instead. Original commit message from CVS: This is an attempt at not segfaulting on errors but reporting some usefull info instead. - bin changes so errors can propagate. - changed the _FAST macros to _CAST because that is what they do. - removed all references to cothreads out of the core, they are really a scheduler issue, handler with a sched_private gpointer. - added a live buffer count, for debugging buffer leaks. - added error checking in gst_scheduler_state_transition this solves the "out of cothreads" problem. - GST_ELEMENT_NO_ENTRY == GST_ELEMENT_INFINITE_LOOP - added 2 private element flasg for use by the scheduler (_COTHREAD_STOPPING) is now - added scheduler entry points: - _yield : to create possible scheduling points. - _interrupt: to stop execution of an element. - _error: to signal en error condition to the scheduler. - improved error messages for pads. - signal gst_element_error where appropriate. - added the a new bin to the parent before entering it so one can reference its children. - queue memleak fixes on dispose. - added possible deadlock detection in queue (turned off be default) - GstBasicScheduler is a real class of its own now, hiding its internal variables. - GST_ELEMENT_IS_COTHREAD_STOPPING is gone. either call explicit _yield operations, or make a sane loop. - Better state change handling in filesrc. Better error reporting/recovery too. - updated core plugins. - detect non decoupled elements on scheduler boundries and error.
2001-12-22 21:18:17 +00:00
}
/**
* gst_scheduler_interrupt:
* @sched: the scheduler
This is an attempt at not segfaulting on errors but reporting some usefull info instead. Original commit message from CVS: This is an attempt at not segfaulting on errors but reporting some usefull info instead. - bin changes so errors can propagate. - changed the _FAST macros to _CAST because that is what they do. - removed all references to cothreads out of the core, they are really a scheduler issue, handler with a sched_private gpointer. - added a live buffer count, for debugging buffer leaks. - added error checking in gst_scheduler_state_transition this solves the "out of cothreads" problem. - GST_ELEMENT_NO_ENTRY == GST_ELEMENT_INFINITE_LOOP - added 2 private element flasg for use by the scheduler (_COTHREAD_STOPPING) is now - added scheduler entry points: - _yield : to create possible scheduling points. - _interrupt: to stop execution of an element. - _error: to signal en error condition to the scheduler. - improved error messages for pads. - signal gst_element_error where appropriate. - added the a new bin to the parent before entering it so one can reference its children. - queue memleak fixes on dispose. - added possible deadlock detection in queue (turned off be default) - GstBasicScheduler is a real class of its own now, hiding its internal variables. - GST_ELEMENT_IS_COTHREAD_STOPPING is gone. either call explicit _yield operations, or make a sane loop. - Better state change handling in filesrc. Better error reporting/recovery too. - updated core plugins. - detect non decoupled elements on scheduler boundries and error.
2001-12-22 21:18:17 +00:00
* @element: the element requesting an interrupt
*
* Tell the scheduler to interrupt execution of this element.
*
* Returns: TRUE if the element should return NULL from the chain/get
* function.
This is an attempt at not segfaulting on errors but reporting some usefull info instead. Original commit message from CVS: This is an attempt at not segfaulting on errors but reporting some usefull info instead. - bin changes so errors can propagate. - changed the _FAST macros to _CAST because that is what they do. - removed all references to cothreads out of the core, they are really a scheduler issue, handler with a sched_private gpointer. - added a live buffer count, for debugging buffer leaks. - added error checking in gst_scheduler_state_transition this solves the "out of cothreads" problem. - GST_ELEMENT_NO_ENTRY == GST_ELEMENT_INFINITE_LOOP - added 2 private element flasg for use by the scheduler (_COTHREAD_STOPPING) is now - added scheduler entry points: - _yield : to create possible scheduling points. - _interrupt: to stop execution of an element. - _error: to signal en error condition to the scheduler. - improved error messages for pads. - signal gst_element_error where appropriate. - added the a new bin to the parent before entering it so one can reference its children. - queue memleak fixes on dispose. - added possible deadlock detection in queue (turned off be default) - GstBasicScheduler is a real class of its own now, hiding its internal variables. - GST_ELEMENT_IS_COTHREAD_STOPPING is gone. either call explicit _yield operations, or make a sane loop. - Better state change handling in filesrc. Better error reporting/recovery too. - updated core plugins. - detect non decoupled elements on scheduler boundries and error.
2001-12-22 21:18:17 +00:00
*/
gboolean
gst_scheduler_interrupt (GstScheduler * sched, GstElement * element)
This is an attempt at not segfaulting on errors but reporting some usefull info instead. Original commit message from CVS: This is an attempt at not segfaulting on errors but reporting some usefull info instead. - bin changes so errors can propagate. - changed the _FAST macros to _CAST because that is what they do. - removed all references to cothreads out of the core, they are really a scheduler issue, handler with a sched_private gpointer. - added a live buffer count, for debugging buffer leaks. - added error checking in gst_scheduler_state_transition this solves the "out of cothreads" problem. - GST_ELEMENT_NO_ENTRY == GST_ELEMENT_INFINITE_LOOP - added 2 private element flasg for use by the scheduler (_COTHREAD_STOPPING) is now - added scheduler entry points: - _yield : to create possible scheduling points. - _interrupt: to stop execution of an element. - _error: to signal en error condition to the scheduler. - improved error messages for pads. - signal gst_element_error where appropriate. - added the a new bin to the parent before entering it so one can reference its children. - queue memleak fixes on dispose. - added possible deadlock detection in queue (turned off be default) - GstBasicScheduler is a real class of its own now, hiding its internal variables. - GST_ELEMENT_IS_COTHREAD_STOPPING is gone. either call explicit _yield operations, or make a sane loop. - Better state change handling in filesrc. Better error reporting/recovery too. - updated core plugins. - detect non decoupled elements on scheduler boundries and error.
2001-12-22 21:18:17 +00:00
{
GstSchedulerClass *sclass;
g_return_val_if_fail (GST_IS_SCHEDULER (sched), FALSE);
g_return_val_if_fail (GST_IS_ELEMENT (element), FALSE);
This is an attempt at not segfaulting on errors but reporting some usefull info instead. Original commit message from CVS: This is an attempt at not segfaulting on errors but reporting some usefull info instead. - bin changes so errors can propagate. - changed the _FAST macros to _CAST because that is what they do. - removed all references to cothreads out of the core, they are really a scheduler issue, handler with a sched_private gpointer. - added a live buffer count, for debugging buffer leaks. - added error checking in gst_scheduler_state_transition this solves the "out of cothreads" problem. - GST_ELEMENT_NO_ENTRY == GST_ELEMENT_INFINITE_LOOP - added 2 private element flasg for use by the scheduler (_COTHREAD_STOPPING) is now - added scheduler entry points: - _yield : to create possible scheduling points. - _interrupt: to stop execution of an element. - _error: to signal en error condition to the scheduler. - improved error messages for pads. - signal gst_element_error where appropriate. - added the a new bin to the parent before entering it so one can reference its children. - queue memleak fixes on dispose. - added possible deadlock detection in queue (turned off be default) - GstBasicScheduler is a real class of its own now, hiding its internal variables. - GST_ELEMENT_IS_COTHREAD_STOPPING is gone. either call explicit _yield operations, or make a sane loop. - Better state change handling in filesrc. Better error reporting/recovery too. - updated core plugins. - detect non decoupled elements on scheduler boundries and error.
2001-12-22 21:18:17 +00:00
sclass = GST_SCHEDULER_GET_CLASS (sched);
if (sclass->interrupt)
return sclass->interrupt (sched, element);
return FALSE;
This is an attempt at not segfaulting on errors but reporting some usefull info instead. Original commit message from CVS: This is an attempt at not segfaulting on errors but reporting some usefull info instead. - bin changes so errors can propagate. - changed the _FAST macros to _CAST because that is what they do. - removed all references to cothreads out of the core, they are really a scheduler issue, handler with a sched_private gpointer. - added a live buffer count, for debugging buffer leaks. - added error checking in gst_scheduler_state_transition this solves the "out of cothreads" problem. - GST_ELEMENT_NO_ENTRY == GST_ELEMENT_INFINITE_LOOP - added 2 private element flasg for use by the scheduler (_COTHREAD_STOPPING) is now - added scheduler entry points: - _yield : to create possible scheduling points. - _interrupt: to stop execution of an element. - _error: to signal en error condition to the scheduler. - improved error messages for pads. - signal gst_element_error where appropriate. - added the a new bin to the parent before entering it so one can reference its children. - queue memleak fixes on dispose. - added possible deadlock detection in queue (turned off be default) - GstBasicScheduler is a real class of its own now, hiding its internal variables. - GST_ELEMENT_IS_COTHREAD_STOPPING is gone. either call explicit _yield operations, or make a sane loop. - Better state change handling in filesrc. Better error reporting/recovery too. - updated core plugins. - detect non decoupled elements on scheduler boundries and error.
2001-12-22 21:18:17 +00:00
}
/**
* gst_scheduler_get_clock:
* @sched: the scheduler
*
* Gets the current clock used by the scheduler.
*
* Returns: a GstClock
*/
GstClock *
gst_scheduler_get_clock (GstScheduler * sched)
{
GstClock *clock = NULL;
/* if we have a fixed clock, use that one */
if (GST_FLAG_IS_SET (sched, GST_SCHEDULER_FLAG_FIXED_CLOCK)) {
clock = sched->clock;
GST_CAT_DEBUG (GST_CAT_CLOCK, "scheduler using fixed clock %p (%s)",
clock, clock ? GST_STR_NULL (GST_OBJECT_NAME (clock)) : "-");
} else {
GList *schedulers = sched->schedulers;
GList *providers = sched->clock_providers;
/* try to get a clock from one of the schedulers we manage first */
while (schedulers) {
GstScheduler *scheduler = GST_SCHEDULER (schedulers->data);
clock = gst_scheduler_get_clock (scheduler);
if (clock) {
GST_CAT_DEBUG (GST_CAT_CLOCK,
"scheduler found managed sched clock %p (%s)",
clock, clock ? GST_STR_NULL (GST_OBJECT_NAME (clock)) : "-");
break;
}
schedulers = g_list_next (schedulers);
}
/* still no clock, try to find one in the providers */
while (!clock && providers) {
clock = gst_element_get_clock (GST_ELEMENT (providers->data));
if (clock)
GST_CAT_DEBUG (GST_CAT_CLOCK, "scheduler found provider clock: %p (%s)",
clock, clock ? GST_STR_NULL (GST_OBJECT_NAME (clock)) : "-");
providers = g_list_next (providers);
}
/* still no clock, use a system clock */
if (!clock && sched->parent_sched == NULL) {
clock = gst_system_clock_obtain ();
/* we unref since this function is not supposed to increase refcount
* of clock object returned; this is ok since the systemclock always
* has a refcount of at least one in the current code. */
gst_object_unref (GST_OBJECT (clock));
GST_CAT_DEBUG (GST_CAT_CLOCK, "scheduler obtained system clock: %p (%s)",
clock, clock ? GST_STR_NULL (GST_OBJECT_NAME (clock)) : "-");
}
}
return clock;
}
/**
* gst_scheduler_use_clock:
* @sched: the scheduler
* @clock: the clock to use
*
* Force the scheduler to use the given clock. The scheduler will
* always use the given clock even if new clock providers are added
* to this scheduler.
*/
void
gst_scheduler_use_clock (GstScheduler * sched, GstClock * clock)
{
g_return_if_fail (sched != NULL);
g_return_if_fail (GST_IS_SCHEDULER (sched));
GST_FLAG_SET (sched, GST_SCHEDULER_FLAG_FIXED_CLOCK);
gst_object_replace ((GstObject **) & sched->clock, (GstObject *) clock);
GST_CAT_DEBUG (GST_CAT_CLOCK, "scheduler using fixed clock %p (%s)", clock,
(clock ? GST_OBJECT_NAME (clock) : "nil"));
}
/**
* gst_scheduler_set_clock:
* @sched: the scheduler
* @clock: the clock to set
*
* Set the clock for the scheduler. The clock will be distributed
* to all the elements managed by the scheduler.
*/
void
gst_scheduler_set_clock (GstScheduler * sched, GstClock * clock)
{
GList *receivers;
GList *schedulers;
g_return_if_fail (sched != NULL);
g_return_if_fail (GST_IS_SCHEDULER (sched));
receivers = sched->clock_receivers;
schedulers = sched->schedulers;
gst_object_replace ((GstObject **) & sched->current_clock,
(GstObject *) clock);
while (receivers) {
GstElement *element = GST_ELEMENT (receivers->data);
GST_CAT_DEBUG (GST_CAT_CLOCK,
"scheduler setting clock %p (%s) on element %s", clock,
(clock ? GST_OBJECT_NAME (clock) : "nil"), GST_ELEMENT_NAME (element));
gst_element_set_clock (element, clock);
receivers = g_list_next (receivers);
}
while (schedulers) {
GstScheduler *scheduler = GST_SCHEDULER (schedulers->data);
GST_CAT_DEBUG (GST_CAT_CLOCK,
"scheduler setting clock %p (%s) on scheduler %p", clock,
(clock ? GST_OBJECT_NAME (clock) : "nil"), scheduler);
gst_scheduler_set_clock (scheduler, clock);
schedulers = g_list_next (schedulers);
}
}
/**
* gst_scheduler_auto_clock:
* @sched: the scheduler
*
* Let the scheduler select a clock automatically.
*/
void
gst_scheduler_auto_clock (GstScheduler * sched)
{
g_return_if_fail (sched != NULL);
g_return_if_fail (GST_IS_SCHEDULER (sched));
GST_FLAG_UNSET (sched, GST_SCHEDULER_FLAG_FIXED_CLOCK);
gst_object_replace ((GstObject **) & sched->clock, NULL);
GST_CAT_DEBUG (GST_CAT_CLOCK, "scheduler using automatic clock");
}
GstClockReturn gst_clock_id_wait (GstClockID id, GstClockTimeDiff * jitter);
/**
* gst_scheduler_clock_wait:
* @sched: the scheduler
* @element: the element that wants to wait
* @id: the clockid to use
* @jitter: the time difference between requested time and actual time
*
* Wait till the clock reaches a specific time. The ClockID can
* be obtained from #gst_clock_new_single_shot_id.
*
* Returns: the status of the operation
*/
GstClockReturn
gst_scheduler_clock_wait (GstScheduler * sched, GstElement * element,
GstClockID id, GstClockTimeDiff * jitter)
{
GstSchedulerClass *sclass;
g_return_val_if_fail (GST_IS_SCHEDULER (sched), GST_CLOCK_ERROR);
g_return_val_if_fail (id != NULL, GST_CLOCK_ERROR);
sclass = GST_SCHEDULER_GET_CLASS (sched);
if (sclass->clock_wait)
return sclass->clock_wait (sched, element, id, jitter);
else
return gst_clock_id_wait (id, jitter);
}
/**
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
* gst_scheduler_iterate:
* @sched: the scheduler
*
* Perform one iteration on the scheduler.
*
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
* Returns: a boolean indicating something usefull has happened.
*/
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
gboolean
gst_scheduler_iterate (GstScheduler * sched)
{
GstSchedulerClass *sclass;
gboolean res = FALSE;
g_return_val_if_fail (GST_IS_SCHEDULER (sched), FALSE);
This is an attempt at not segfaulting on errors but reporting some usefull info instead. Original commit message from CVS: This is an attempt at not segfaulting on errors but reporting some usefull info instead. - bin changes so errors can propagate. - changed the _FAST macros to _CAST because that is what they do. - removed all references to cothreads out of the core, they are really a scheduler issue, handler with a sched_private gpointer. - added a live buffer count, for debugging buffer leaks. - added error checking in gst_scheduler_state_transition this solves the "out of cothreads" problem. - GST_ELEMENT_NO_ENTRY == GST_ELEMENT_INFINITE_LOOP - added 2 private element flasg for use by the scheduler (_COTHREAD_STOPPING) is now - added scheduler entry points: - _yield : to create possible scheduling points. - _interrupt: to stop execution of an element. - _error: to signal en error condition to the scheduler. - improved error messages for pads. - signal gst_element_error where appropriate. - added the a new bin to the parent before entering it so one can reference its children. - queue memleak fixes on dispose. - added possible deadlock detection in queue (turned off be default) - GstBasicScheduler is a real class of its own now, hiding its internal variables. - GST_ELEMENT_IS_COTHREAD_STOPPING is gone. either call explicit _yield operations, or make a sane loop. - Better state change handling in filesrc. Better error reporting/recovery too. - updated core plugins. - detect non decoupled elements on scheduler boundries and error.
2001-12-22 21:18:17 +00:00
sclass = GST_SCHEDULER_GET_CLASS (sched);
if (sclass->iterate) {
res = sclass->iterate (sched);
}
return res;
}
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
/**
* gst_scheduler_show:
* @sched: the scheduler
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
*
* Dump the state of the scheduler
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
*/
void
gst_scheduler_show (GstScheduler * sched)
{
GstSchedulerClass *sclass;
This is an attempt at not segfaulting on errors but reporting some usefull info instead. Original commit message from CVS: This is an attempt at not segfaulting on errors but reporting some usefull info instead. - bin changes so errors can propagate. - changed the _FAST macros to _CAST because that is what they do. - removed all references to cothreads out of the core, they are really a scheduler issue, handler with a sched_private gpointer. - added a live buffer count, for debugging buffer leaks. - added error checking in gst_scheduler_state_transition this solves the "out of cothreads" problem. - GST_ELEMENT_NO_ENTRY == GST_ELEMENT_INFINITE_LOOP - added 2 private element flasg for use by the scheduler (_COTHREAD_STOPPING) is now - added scheduler entry points: - _yield : to create possible scheduling points. - _interrupt: to stop execution of an element. - _error: to signal en error condition to the scheduler. - improved error messages for pads. - signal gst_element_error where appropriate. - added the a new bin to the parent before entering it so one can reference its children. - queue memleak fixes on dispose. - added possible deadlock detection in queue (turned off be default) - GstBasicScheduler is a real class of its own now, hiding its internal variables. - GST_ELEMENT_IS_COTHREAD_STOPPING is gone. either call explicit _yield operations, or make a sane loop. - Better state change handling in filesrc. Better error reporting/recovery too. - updated core plugins. - detect non decoupled elements on scheduler boundries and error.
2001-12-22 21:18:17 +00:00
g_return_if_fail (GST_IS_SCHEDULER (sched));
sclass = GST_SCHEDULER_GET_CLASS (sched);
if (sclass->show)
sclass->show (sched);
}
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
/*
* Factory stuff starts here
*
*/
static void gst_scheduler_factory_class_init (GstSchedulerFactoryClass * klass);
static void gst_scheduler_factory_init (GstSchedulerFactory * factory);
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
static GstPluginFeatureClass *factory_parent_class = NULL;
/* static guint gst_scheduler_factory_signals[LAST_SIGNAL] = { 0 }; */
GType
gst_scheduler_factory_get_type (void)
{
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
static GType schedulerfactory_type = 0;
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
if (!schedulerfactory_type) {
static const GTypeInfo schedulerfactory_info = {
sizeof (GstSchedulerFactoryClass),
NULL,
NULL,
(GClassInitFunc) gst_scheduler_factory_class_init,
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
NULL,
NULL,
sizeof (GstSchedulerFactory),
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
0,
(GInstanceInitFunc) gst_scheduler_factory_init,
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
NULL
};
schedulerfactory_type = g_type_register_static (GST_TYPE_PLUGIN_FEATURE,
"GstSchedulerFactory", &schedulerfactory_info, 0);
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
}
return schedulerfactory_type;
}
static void
gst_scheduler_factory_class_init (GstSchedulerFactoryClass * klass)
{
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
GObjectClass *gobject_class;
GstObjectClass *gstobject_class;
GstPluginFeatureClass *gstpluginfeature_class;
gobject_class = (GObjectClass *) klass;
gstobject_class = (GstObjectClass *) klass;
gstpluginfeature_class = (GstPluginFeatureClass *) klass;
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
factory_parent_class = g_type_class_ref (GST_TYPE_PLUGIN_FEATURE);
if (!_default_name) {
if (g_getenv ("GST_SCHEDULER")) {
_default_name = g_strdup (g_getenv ("GST_SCHEDULER"));
} else {
_default_name = g_strdup (GST_SCHEDULER_DEFAULT_NAME);
}
}
g_assert (_default_name);
}
static void
gst_scheduler_factory_init (GstSchedulerFactory * factory)
{
}
/**
* gst_scheduler_register:
* @plugin: a #GstPlugin
* @name: name of the scheduler to register
* @longdesc: description of the scheduler
* @type: #GType of the scheduler to register
*
* Registers a scheduler with GStreamer.
*
* Returns: TRUE, if the registering succeeded, FALSE on error.
*
* Since: 0.8.5
**/
gboolean
gst_scheduler_register (GstPlugin * plugin, const gchar * name,
const gchar * longdesc, GType type)
{
GstSchedulerFactory *factory;
g_return_val_if_fail (plugin != NULL, FALSE);
g_return_val_if_fail (name != NULL, FALSE);
g_return_val_if_fail (longdesc != NULL, FALSE);
g_return_val_if_fail (g_type_is_a (type, GST_TYPE_SCHEDULER), FALSE);
factory = gst_scheduler_factory_find (name);
if (factory) {
g_return_val_if_fail (factory->type == 0, FALSE);
g_free (factory->longdesc);
factory->longdesc = g_strdup (longdesc);
factory->type = type;
} else {
factory = gst_scheduler_factory_new (name, longdesc, type);
g_return_val_if_fail (factory, FALSE);
gst_plugin_add_feature (plugin, GST_PLUGIN_FEATURE (factory));
}
return TRUE;
}
/**
* gst_scheduler_factory_new:
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
* @name: name of schedulerfactory to create
* @longdesc: long description of schedulerfactory to create
* @type: the gtk type of the GstScheduler element of this factory
*
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
* Create a new schedulerfactory with the given parameters
*
* Returns: a new #GstSchedulerFactory.
*/
GstSchedulerFactory *
gst_scheduler_factory_new (const gchar * name, const gchar * longdesc,
GType type)
{
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
GstSchedulerFactory *factory;
g_return_val_if_fail (name != NULL, NULL);
factory = gst_scheduler_factory_find (name);
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
if (!factory) {
factory =
GST_SCHEDULER_FACTORY (g_object_new (GST_TYPE_SCHEDULER_FACTORY, NULL));
GST_PLUGIN_FEATURE_NAME (factory) = g_strdup (name);
} else {
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
g_free (factory->longdesc);
}
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
factory->longdesc = g_strdup (longdesc);
factory->type = type;
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
return factory;
}
/**
* gst_scheduler_factory_destroy:
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
* @factory: factory to destroy
*
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
* Removes the scheduler from the global list.
*/
void
gst_scheduler_factory_destroy (GstSchedulerFactory * factory)
{
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
g_return_if_fail (factory != NULL);
/* we don't free the struct bacause someone might have a handle to it.. */
}
/**
* gst_scheduler_factory_find:
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
* @name: name of schedulerfactory to find
*
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
* Search for an schedulerfactory of the given name.
*
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
* Returns: #GstSchedulerFactory if found, NULL otherwise
*/
GstSchedulerFactory *
gst_scheduler_factory_find (const gchar * name)
{
GstPluginFeature *feature;
g_return_val_if_fail (name != NULL, NULL);
GST_DEBUG ("gstscheduler: find \"%s\"", name);
feature = gst_registry_pool_find_feature (name, GST_TYPE_SCHEDULER_FACTORY);
if (feature)
return GST_SCHEDULER_FACTORY (feature);
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
return NULL;
}
/**
* gst_scheduler_factory_create:
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
* @factory: the factory used to create the instance
* @parent: the parent element of this scheduler
*
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
* Create a new #GstScheduler instance from the
* given schedulerfactory with the given parent. @parent will
* have its scheduler set to the returned #GstScheduler instance.
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
*
* Returns: A new #GstScheduler instance with a reference count of %1.
*/
GstScheduler *
gst_scheduler_factory_create (GstSchedulerFactory * factory,
GstElement * parent)
{
GstScheduler *sched = NULL;
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
g_return_val_if_fail (factory != NULL, NULL);
g_return_val_if_fail (GST_IS_ELEMENT (parent), NULL);
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
if (gst_plugin_feature_ensure_loaded (GST_PLUGIN_FEATURE (factory))) {
g_return_val_if_fail (factory->type != 0, NULL);
sched = GST_SCHEDULER (g_object_new (factory->type, NULL));
sched->parent = parent;
GST_ELEMENT_SCHED (parent) = sched;
/* let's refcount the scheduler */
gst_object_ref (GST_OBJECT (sched));
gst_object_sink (GST_OBJECT (sched));
}
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
return sched;
}
/**
* gst_scheduler_factory_make:
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
* @name: the name of the factory used to create the instance
* @parent: the parent element of this scheduler
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
*
* Create a new #GstScheduler instance from the
* schedulerfactory with the given name and parent. @parent will
* have its scheduler set to the returned #GstScheduler instance.
* If %NULL is passed as @name, the default scheduler name will
* be used.
*
* Returns: A new #GstScheduler instance with a reference count of %1.
*/
GstScheduler *
gst_scheduler_factory_make (const gchar * name, GstElement * parent)
{
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
GstSchedulerFactory *factory;
const gchar *default_name = gst_scheduler_factory_get_default_name ();
if (name)
factory = gst_scheduler_factory_find (name);
else {
/* FIXME: do better error handling */
if (default_name == NULL)
g_error ("No default scheduler name - do you have a registry ?");
factory = gst_scheduler_factory_find (default_name);
}
- some fixes to int2float making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it bui... Original commit message from CVS: - added playondemand plugin by Leif Morgan Johnson <lmjohns3@eos.ncsu.edu> - some fixes to int2float - aplied a patch from wrobell <wrobell@ite.pl> that is a first attempt at making automake 1.5 happy (gst now requires automake1.5). It's still not perfect but it builds. - Made the schedulers plugable. The default scheduler now lives inside a plugin. - Added a new mpeg1/2 parser/demuxer. - Fixed some compiler warnings in the core libs. - substantial work to GstThread (hopefully less race conditions). simplified the code in GstThread a bit. A state change can now also happen in the thread context. - reworked the state semantics of a bin. it'll now automatically get the highest state of its children. - the autoplugger now nests the threads so that a state change failure of one thread doesn't make its upstream thread lock. - GstQueue refuses to go to PLAYING if the sinkpad is not connected. This way the queue will not wedge in the _get lock. - GstQueue unlocks its mutexes when going to PAUSED. - make sure that when all elements in a bin/thread go to PAUSED, the bin is set to PAUSED too. - make a parent bin wait for its children to PAUSE before ending the iteration with FALSE (EOS) - Some changes to GstPlay to deal with EOS. - aplied the latest patch from Zeenix to gstrtp. end result: GstPlay doesn't crash on EOS and the pipeline is now shut down properly.
2001-12-04 22:12:50 +00:00
if (factory == NULL)
return NULL;
return gst_scheduler_factory_create (factory, parent);
}
/**
* gst_scheduler_factory_set_default_name:
* @name: the name of the factory used as a default
*
* Set the default schedulerfactory name.
*/
void
gst_scheduler_factory_set_default_name (const gchar * name)
{
g_free (_default_name);
_default_name = g_strdup (name);
}
/**
* gst_scheduler_factory_get_default_name:
*
* Get the default schedulerfactory name.
*
* Returns: the name of the default scheduler.
*/
const gchar *
gst_scheduler_factory_get_default_name (void)
{
return _default_name;
}