Add private replacements for deprecated functions such as
g_mutex_new(), g_mutex_free(), g_cond_new() etc., mostly
to avoid the deprecation warnings. We can't change most of
these in 0.10 because they're part of our API and ABI.
Drop in old GstBus code for the release to play it safe, since
regressions that are apparently hard to track down and reproduce
have been reported (on windows/OSX mostly) against the lockfree
version, and more time is needed to fix them.
This reverts commit 03391a8970.
This reverts commit 43cdbc17e6.
This reverts commit 80eb160e0f.
This reverts commit c41b0ade28.
This reverts commit 874d60e589.
This reverts commit 79370d4b17.
This reverts commit 2cb3e52351.
This reverts commit bd1c400114.
This reverts commit 4bf8f1524f.
This reverts commit 14d7db1b52.
https://bugzilla.gnome.org/show_bug.cgi?id=647493
... which happens in particular flushing a bus, possibly as part
of a state change, e.g. when having a pipeline in a pipeline
and then changing state back to NULL. The interior pipeline
will/might then flush the bus, which is a child bus from the
parent which does not have a poll anymore these days.
https://bugzilla.gnome.org/show_bug.cgi?id=648297
This allows to only create the socketpair when it is really required instead
of always creating it and immediately destroying it again for child buses.
https://bugzilla.gnome.org/show_bug.cgi?id=647005
This is used by GstBin to create a child bus without
a socketpair because child buses will always work
synchronous. Otherwise too many sockets could be
created and the limit of file descriptors for the
process could be reached.
Fixes bug #646624.
Use new GstPoll functionality to wakeup the mainloop.
Use an atomic queue on the writer side to post the messages.
The reader side it protected with the lock still because we don't want multiple
concurrent readers.
Makes gst_bus_add_watch(), gst_bus_add_watch_full(), gst_bus_add_signal_watch(),
and gst_bus_add_signal_watch_full() convenience functions automatically pick up
any non-default main contexts set for the current thread via
g_main_thread_push_thread_default().
This either must never happen (which makes sense in this case) and thus should
use assert() or we should use a traditional if (poll_data->message) return;
to avoid differnet behaviour of intenal api when compiling with
G_DISABLE_CHECKS.
Adds that warning to configure.ac
Includes a tiny change of the GST_BOILERPLATE_FULL() macro:
The get_type() function is no longer declared before being defined.
https://bugzilla.gnome.org/show_bug.cgi?id=611692
This avoids:
* triple-checking for the GType when type-checking is enabled (see #597260)
* Avoids going through an expensive no-argument checking which landed in
glib-2.22
* Avoids going through 2 extrac functions (g_object_new -> g_object_new_valist)
There's not much point in using GST_DEBUG_FUNCPTR with GObject
virtual functions such as get_property, set_propery, finalize and
dispose, since they'll never be used by anyone anyway. Saves a
few bytes and possibly a tenth of a polar bear.
Original commit message from CVS:
* gst/gstbus.c: (gst_bus_set_main_context), (gst_bus_create_watch):
Fix order of members in GstBusSource structure - the first member
must be the parent structure ie. GSource. Should make bus sources
attached to non-default main contexts work in all cases now (ie.
primarily in cases where the callback has a non-NULL user data
argument). Fixes#562170.
* tests/check/gst/gstbus.c: (test_custom_main_context):
Add unit test for the above, based on code by
Justin Karneges <justin at affinix com>.
Original commit message from CVS:
* gst/gstbus.c: (gst_bus_wakeup_main_context):
The lock order should be maincontext > OBJECT_LOCK so we need to release
the object lock when waking up the mainloop to avoid deadlocks.
Original commit message from CVS:
* gst/gstbus.c: (gst_bus_dispose), (gst_bus_get_property),
(gst_bus_wakeup_main_context), (gst_bus_set_main_context),
(gst_bus_post), (gst_bus_source_prepare), (gst_bus_source_finalize),
(gst_bus_create_watch):
Make GstBusSource work with non-default main contexts (#562170).
* tests/check/gst/gstbus.c: (message_func_eos), (message_func_app),
(test_watch), (test_watch_with_custom_context), (gst_bus_suite):
Add test case for GstBusSource with a non-default main context.
* tests/check/libs/.cvsignore:
Ignore more.
Original commit message from CVS:
* gst/gstbus.c: (gst_bus_source_finalize),
(gst_bus_add_watch_full_unlocked), (gst_bus_add_watch_full),
(gst_bus_enable_sync_message_emission),
(gst_bus_disable_sync_message_emission),
(gst_bus_add_signal_watch_full), (gst_bus_remove_signal_watch):
Fix deadlock, g_source_get_id() cannot be called in finalize.
Keep track of the watch source by keeping a pointer to the source object
instead.
Use the bus lock to protect access to the pointer to the current
watch source.
Original commit message from CVS:
Base on Patch by: Olivier Crete <tester at tester dot ca>
* gst/gstbus.c: (gst_bus_source_finalize),
(gst_bus_add_watch_full), (gst_bus_add_signal_watch_full):
Only allow one bus watch to be set at a time. This is necessary
because the dispatcher pops the message from the bus and the second
watcher will then get NULL or the next message (and the first won't
get this next message then, etc). If more than one "watcher" is
required signal watches should be used. Fixes bug #526044.
Original commit message from CVS:
Patch by: Sjoerd Simons <sjoerd at luon net>
* gst/gstbus.c: (gst_bus_source_dispatch):
Don't print a warning if the queue is empty when we try to pop
here. That could happen if another thread or callback set the
bus to flushing between the source's check/prepare and the
dispatch being called (#531538).
Original commit message from CVS:
* docs/gst/gstreamer-sections.txt:
* gst/gstbus.c:
* gst/gstbus.h:
API: add gst_bus_pop_filtered
API: add gst_bus_timed_pop_filtered
Two new functions for waiting for specific message types on the
bus for a specified amount of time without iterating any main
loops or main contexts.
* tests/check/gst/gstbus.c:
Some tests for the new functions.
Original commit message from CVS:
* docs/gst/gstreamer-sections.txt:
* gst/gstbus.c: (gst_bus_init), (gst_bus_dispose), (gst_bus_post),
(gst_bus_timed_pop), (gst_bus_pop):
* gst/gstbus.h:
API: gst_bus_timed_pop()
Implement gst_bus_timed_pop() to do a blocking timed wait for a
message to arrive on the bus.
* tests/check/gst/gstbus.c: (GST_START_TEST), (pop_thread),
(gst_bus_suite):
Two unit tests for new _timed_pop() function.
Original commit message from CVS:
* gst/gstbus.c: (gst_bus_class_init), (gst_bus_post):
More docs for the sync-message signal (mention that it is not
emitted by default); log message structures of messages posted on
the bus as well.
Original commit message from CVS:
Patch by: James Andrewartha <trs80 at tartarus uwa edu au>
* gst/gst.c:
* gst/gstbus.c:
* gst/gstclock.c:
* gst/gstevent.c:
* gst/gstformat.c:
* gst/gstmessage.c:
* gst/gstparse.c:
* gst/gstquery.c:
* gst/gstutils.c:
* gst/parse/Makefile.am:
* libs/gst/base/gstadapter.c:
* libs/gst/base/gstbasesrc.c:
* libs/gst/base/gstpushsrc.c:
* libs/gst/base/gsttypefindhelper.c:
* plugins/elements/gstfakesrc.c:
* plugins/elements/gstidentity.c:
Make sure gstprivate.h and/or config.h are
always included first, otherwise some of our
defines (like _FILE_OFFSET_BITS) might be
redefined in the system headers. Fixes build
on opensolaris (#340016).
Original commit message from CVS:
* gst/gstbus.c:
Small documentation clarification about the signal watch.
* libs/gst/base/gstbasesink.c: (gst_base_sink_get_sync_times),
(gst_base_sink_wait_clock), (gst_base_sink_do_sync),
(gst_base_sink_perform_qos), (gst_base_sink_reset_qos),
(gst_base_sink_do_render_stats), (gst_base_sink_render_object),
(gst_base_sink_get_position_last),
(gst_base_sink_get_position_paused), (gst_base_sink_change_state):
Convert and store timestamps in stream time and running time, the
raw timestamps are not usefull, also document this better.
Use different window sizes for good and bad QoS observations so
we react to badness a little quicker.
Keep track of the amount of rendered and dropped buffers.
Send QoS timestamps in running time.
* libs/gst/base/gstbasetransform.c:
(gst_base_transform_sink_eventfunc),
(gst_base_transform_handle_buffer):
Compare QoS timestamps against running time.
Original commit message from CVS:
* gst/gstbin.c: (gst_bin_get_type), (gst_bin_handle_message_func):
Help the compiler a bit with type registration.
Use existing forward cod path instead of duplicating it when
handling a message.
* gst/gstbus.c: (gst_bus_get_type):
* gst/gstcaps.c: (gst_caps_get_type), (gst_static_caps_get_type):
* gst/gstchildproxy.c: (gst_child_proxy_get_type):
* gst/gstclock.c: (gst_clock_get_type):
* gst/gstelement.c: (gst_element_get_type),
* gst/gstelementfactory.c: (gst_element_factory_get_type):
* gst/gstindexfactory.c: (gst_index_factory_get_type):
* gst/gstminiobject.c: (gst_mini_object_get_type):
* gst/gstpad.c: (gst_pad_get_type):
* gst/gstsegment.c: (gst_segment_get_type):
* gst/gststructure.c: (gst_structure_get_type):
* gst/gstsystemclock.c: (gst_system_clock_get_type):
* gst/gsttask.c: (gst_task_get_type), (gst_task_join):
* gst/gstvalue.c:
Help compiler with type registration.
* plugins/elements/gstqueue.c: (gst_queue_handle_sink_event):
Small doc update.
Original commit message from CVS:
2006-02-10 Andy Wingo <wingo@pobox.com>
* gst/gstbus.c (gst_bus_class_init): Declare our private data
structure.
(gst_bus_init): Cache the location of the private data in the
instance structure.
(gst_bus_enable_sync_message_emission)
(gst_bus_disable_sync_message_emission): Implement new public
functions.
(gst_bus_post): Emit the sync-message signal if the user asked for
it. Fixes#330684.
* gst/gstbus.h (GstBus): Use a padding pointer to cache the
location of the bus-private structuure.
(gst_bus_enable_sync_message_emission)
(gst_bus_disable_sync_message_emission): New public functions.
Original commit message from CVS:
* gst/base/gstbasesink.c: (gst_base_sink_handle_object):
Make sure the GstFlowReturn is returned.
* gst/gstbus.c: (gst_bus_add_signal_watch_full),
(gst_bus_add_signal_watch):
* gst/gstbus.h:
add gst_bus_add_signal_watch_full.
* gst/gstplugin.c: (gst_plugin_load_file):
Small style cleanup.
Original commit message from CVS:
2005-11-21 Andy Wingo <wingo@pobox.com>
* *.h:
* *.c: Ran scripts/update-macros. Oh yes.
* gst/gstobject.h (GST_OBJECT_GET_LOCK, GST_OBJECT_LOCK)
(GST_OBJECT_TRYLOCK, GST_OBJECT_UNLOCK): Renamed from
GST_GET_LOCK, etc.
* scripts/update-macros: New script. Run it on your files to
change GST_LOCK to GST_OBJECT_LOCK, and the same for UNLOCK as
well.
Original commit message from CVS:
2005-11-10 Andy Wingo <wingo@pobox.com>
* check/gst/gstbin.c (test_message_state_changed_children): Style
fix..
* gst/gstbus.c (poll_destroy, poll_func, gst_bus_poll): Implement
gst_bus_poll with the signal watch. Ensures that poll and a signal
watch see the same messages.
* check/gst/gstbus.c (test_watch_with_poll): New test, checks that
a poll and a watch at the same time get the same messages.
Original commit message from CVS:
* gst/gstbus.c: (gst_bus_post), (gst_bus_set_sync_handler):
Don't leak a mutex unlock in case of an error.
* gst/gstbus.h:
Doc fixes.
Original commit message from CVS:
* gst/gstbus.c: (gst_bus_class_init):
Fix some typos.
* gst/gstqueue.c: (gst_queue_loop):
Don't assume a miniobject that isn't a buffer is an
event (it could be that there is a refcounting
problem somewhere and the pointer is stale and
refers to an already destroyed miniobject).
Original commit message from CVS:
* gst/gstbus.c:
Doc fix.
* gst/gstqueue.c: (gst_queue_handle_sink_event), (gst_queue_chain),
(gst_queue_loop):
Fix potential deadlock when QUEUE_LOCK is taken before STREAM_LOCK.
Original commit message from CVS:
* gst/gstbus.c: (gst_bus_pop):
* gst/gstclock.c:
* gst/gstsystemclock.c: (gst_system_clock_async_thread):
Small cleanups and doc updates.
Original commit message from CVS:
* gst/gst_private.h:
* gst/gstbus.c:
* gst/gstelement.c:
* gst/gstinfo.c:
* gst/gstpluginfeature.c:
widen the debug category in output to fit the biggest one we have
add a bus category and use it
play with the colors
fix up some categories
Original commit message from CVS:
2005-09-29 Andy Wingo <wingo@pobox.com>
* gst/gstbus.c (gst_bus_remove_signal_watch): New function,
removes signal watches previously added via
gst_bus_add_signal_watch.
(gst_bus_add_signal_watch): Don't return the source id, just store
it on the bus if there wasn't an id already.
* gst/gstbus.h (GstBus): Add a couple new fields. API changes for
add_signal_watch and remove_signal_watch.
Original commit message from CVS:
* check/gst/gstbin.c: (GST_START_TEST):
Change for new bus API.
* check/gst/gstbus.c: (message_func_eos), (message_func_app),
(send_messages), (GST_START_TEST), (gstbus_suite):
Change for new bus signal API.
* gst/gstbus.c: (gst_bus_class_init), (gst_bus_have_pending),
(gst_bus_source_prepare), (gst_bus_source_check),
(gst_bus_create_watch), (gst_bus_add_watch_full),
(gst_bus_add_watch), (gst_bus_poll), (gst_bus_async_signal_func),
(gst_bus_sync_signal_handler), (gst_bus_add_signal_watch):
* gst/gstbus.h:
Remove support for multiple GSources operating on different
message types as it is too complex and unneeded when using
signals.
Added support for receiving signals from the bus.
Original commit message from CVS:
* gst/gstbus.c: (marshal_VOID__MINIOBJECT), (gst_bus_class_init),
(gst_bus_post), (poll_func), (gst_bus_async_signal_func),
(gst_bus_sync_signal_handler):
* gst/gstbus.h:
Added async-message and sync-message signals to the bus.
Added helper BusFunc to emit signals for all posted messages.
* gst/gstmessage.c: (gst_message_type_get_name),
(gst_message_type_to_quark), (gst_message_get_type):
* gst/gstmessage.h:
Register quarks for message names.
Original commit message from CVS:
* check/gst/gstpipeline.c: (GST_START_TEST), (gst_pipeline_suite):
add another check
* gst/gstbus.c:
add some doc
* gst/gstinfo.c: (_gst_debug_init):
slightly more readable color for refcount debugging
Original commit message from CVS:
* docs/design/part-gstbin.txt:
* docs/design/part-gstbus.txt:
* gst/gstbus.c:
Add blurb about how the bus goes into flushing mode and
drops all messages when its bin goes from READY into NULL
state.
Original commit message from CVS:
* check/gst/gstbin.c: (test_children_state_change_order_flagged_sink),
(test_children_state_change_order_semi_sink), (gst_bin_suite):
Added test to check state change order in bins (can still be made
to fail here under heavy disk load; bails out with 'Push on pad
fakesink:sink0, but it was not activated in push mode').
* gst/gstbin.c: (gst_bin_class_init), (gst_bin_change_state):
Fix state change order when there is only a semi sink (#316856)
* gst/gstbus.c: (gst_bus_class_init):
Use _class_peek_parent(), not _class_ref(); fix docs to say
'default main context' instead of 'mainloop' where that is
what's meant.
* gst/gstelement.c: (gst_element_commit_state),
(gst_element_set_state):
Fix typos in debug messages
Original commit message from CVS:
* gst/gstbus.c: (poll_func), (poll_timeout), (gst_bus_poll):
* tools/gst-launch.c: (main):
The timeout handler should have lower priority than the source
so we don't timeout before popping a message with 0 timeout.
Dump error messages after failed state change.
Original commit message from CVS:
* check/gst/gstbin.c: (pop_messages), (GST_START_TEST):
* check/gst/gstbus.c: (message_func_eos), (message_func_app),
(send_messages), (GST_START_TEST), (gstbus_suite):
* check/gst/gstpipeline.c: (GST_START_TEST):
* check/pipelines/cleanup.c: (run_pipeline):
* check/pipelines/simple_launch_lines.c: (run_pipeline),
(GST_START_TEST):
* gst/gstbus.c: (gst_bus_have_pending), (gst_bus_source_prepare),
(gst_bus_source_check), (gst_bus_source_dispatch),
(gst_bus_create_watch), (gst_bus_add_watch_full),
(gst_bus_add_watch), (poll_func), (poll_timeout), (gst_bus_poll):
* gst/gstbus.h:
* tools/gst-launch.c: (event_loop):
* tools/gst-md5sum.c: (event_loop):
GstBusHandler -> GstBusFunc, return value has the same meaning as
any other GSource (FALSE == remove source).
_add_watch() and _add_watch_full() now take a MessageType mask to
only handle specific types of messages.
_poll() returns the GstMessage instead of the message type to avoid
race conditions.
_have_pending() takes a MessageType mask now too.
Added testsuite for multiple bus watches.
Fix testsuites and applications for new bus API.
