Commit graph

183 commits

Author SHA1 Message Date
Matthew Waters
6f9a63a10d gst/systemclock: wait on each entry individually
Problem:
multiple aggregator elements (audiomixer, compositor) in a live
pipeline use a lot of CPU waiting each other up.  This is because
of the previously unused clock entry unscheduling during regular
operation.

Clock entry unscheduling has the potential to wake up every clock entry
waiting using the system clock which may be a large number.

Solution:
Implement waiting per entry and only wakeup the unscheduled entry.

While this may be possible using GCond, theoretically GCond only gives
us microsecond accuracy and uses relative waits in a number of places.
We can unfortunately do better poking at the platform specifics
ourselves by using futexes on linux and pthread on other unix.  Windows
may have a possible implementation using Waitable timers but that is
not implemented here and instead falls back to the GCond implementation.
GCond waits on Windows is still as accurate as the previous GstPoll-based
implementation.
2020-04-16 01:26:59 +00:00
Niels De Graef
4812c4087f Don't pass default GLib marshallers for signals
By passing NULL to `g_signal_new` instead of a marshaller, GLib will
actually internally optimize the signal (if the marshaller is available
in GLib itself) by also setting the valist marshaller. This makes the
signal emission a bit more performant than the regular marshalling,
which still needs to box into `GValue` and call libffi in case of a
generic marshaller.

Note that for custom marshallers, one would use
`g_signal_set_va_marshaller()` with the valist marshaller instead.
2019-09-09 12:16:24 +00:00
Sebastian Dröge
1fda8c3bcf clock: Move clock GWeakRef to a private GstClockEntry struct
There's no need for it to be in the public struct and we can keep the
padding for things to be added in the future.
2018-11-08 09:07:07 +00:00
Sebastian Dröge
8250b8d81a clock: Move clock weak ref into its own ABI struct
Otherwise it will be hard to add other things into the padding later
without breaking API.
2018-11-03 19:00:36 +02:00
Sebastian Dröge
4de89865d4 clock: Add new functions to the documentation 2018-11-03 19:00:35 +02:00
Sebastian Dröge
f34472822c clock: Fix deprecation handling of the GstClock clock field 2018-11-03 19:00:24 +02:00
Thomas Bluemel
36ab067905 clock: Keep weak reference to underlying clock
Fixes potential segmentation fault when using a GstClockID that
is referencing an already freed GstClock

Fixes https://gitlab.freedesktop.org/gstreamer/gstreamer/issues/187
2018-11-03 19:00:22 +02:00
Tim-Philipp Müller
2db8e3705f Update for g_type_class_add_private() deprecation in recent GLib
https://gitlab.gnome.org/GNOME/glib/merge_requests/7
2018-06-24 12:49:14 +02:00
Sebastian Dröge
daa98fc02a gst: Don't ref_sink() GstObject subclasses in instance_init/constructor
This is something bindings can't handle and it causes leaks. Instead
move the ref_sink() to the explicit, new() constructors.

This means that abstract classes, and anything that can have subclasses,
will have to do ref_sink() in their new() function now. Specifically
this affects GstClock and GstControlSource.

https://bugzilla.gnome.org/show_bug.cgi?id=743062
2017-05-17 10:40:37 +03:00
Thibault Saunier
a87b4551a6 Port gtk-doc comments to their equivalent markdown syntax
Modernizing our documentation and preparing a possible move to hotdoc.
This commits also adds missing @title metadatas to all SECTIONs
2017-01-27 16:36:38 -03:00
Sebastian Dröge
160ca37139 clock: Fix offsetting of times_temp relative to the times array 2016-11-29 18:14:58 +02:00
Sebastian Dröge
a7d282d272 utils: Export linear regression calculation as public function
It is useful outside the GstClock code too.

https://bugzilla.gnome.org/show_bug.cgi?id=774916
2016-11-28 14:26:50 +02:00
Tim-Philipp Müller
e2cd6ffafc Remove old alloc tracing code now that we have a GstTracer-based replacement
It's been internal API only in 1.x.
2016-08-13 10:10:30 +01:00
Sebastian Dröge
c32ea3cb58 clock: Return FALSE in all paths that don't set out parameters in gst_clock_add_observation_unapplied()
It returned TRUE when regression failed, while not setting any of the out
parameters. This caused uninitialized data from the stack to be used for
setting the clock calibration.
2016-04-02 01:34:46 +03:00
Sebastian Dröge
5fe1b7efe3 clock: adjust/unadjust_with_calibration() have a clock parameter but it's useless
Document this, for 2.0 we should just remove that parameter.
2016-01-06 16:25:38 +02:00
Sebastian Dröge
fb01dd000b clock: Add gst_clock_unadjust_with_calibration()
We already have gst_clock_adjust_with_calibration() and
gst_clock_unadjust_unlocked(), having the other variant is useful.
2016-01-06 16:25:38 +02:00
Sebastian Dröge
01ba6d40a8 clock: Fix typo
clocked -> clock
2016-01-05 16:44:53 +02:00
Sebastian Dröge
2f776d679f clock: Don't allow setting an unsynced clock as master 2016-01-05 14:23:26 +02:00
Sebastian Dröge
ab827eca2e clock: Don't try to slave unsynced clocks
They will return useless values from get_time().
2016-01-05 14:21:58 +02:00
Nicolas Dufresne
b059520926 doc: Fix typo in ref _clock_wait_for_sync() 2015-06-13 20:52:01 -04:00
Sebastian Dröge
fe3249b0e1 clock: Add GST_CLOCK_FLAG_NEEDS_STARTUP_SYNC and related API
gst_clock_wait_for_sync(), gst_clock_is_synced() and gst_clock_set_synced()
plus a signal to asynchronously wait for the clock to be synced.

This can be used by clocks to signal that they need initial synchronization
before they can report any time, and that this synchronization can also get
completely lost at some point. Network clocks, like the GStreamer
netclientclock, NTP or PTP clocks are examples for clocks where this is useful
to have as they can't report any time at all before they're synced.

https://bugzilla.gnome.org/show_bug.cgi?id=749391
2015-06-03 13:53:27 +02:00
Jan Schmidt
81d613694c clock: Improve slaving regression.
Add domain checks for the input values, and a variable precision
calculation that loops if necessary to ensure we never overflow
accumulators and then silently produce garbage results.

Make the (non-public) linear regression function available for
unit testing by putting it in a separate source file the test
can include. Add a unit test that the new regression function
produces sensible results for several inputs taken from real-world
captures.
2015-01-21 22:27:18 +11:00
Jan Schmidt
4bc7d34f57 clock: Add gst_clock_add_observation_unapplied()
gst_clock_add_observation_unapplied() adds a new master/slave clock
observation and runs the regression without activating the new
calibration results.
2015-01-21 22:27:18 +11:00
Jan Schmidt
b79b14815e clock: Add gst_clock_adjust_with_calibration()
gst_clock_adjust_with_calibration() uses directly passed calibration
parameters, instead of using the clock's current calibration,
allowing for calculations using pending or old calibration params
2015-01-21 22:27:18 +11:00
Evan Nemerson
2759882379 introspection: add missing (nullable) annotations to return values
Support for (nullable) was added to G-I at the same time as nullable
return values.  Previous versions of G-I will not mark return values as
nullable, even when an (allow-none) annotation is present, so it is
not necessary to add (allow-none) annotations for compatibility with
older versions of G-I.

https://bugzilla.gnome.org/show_bug.cgi?id=730957
2014-06-26 18:56:38 +02:00
Tim-Philipp Müller
6eb6d9ec38 docs: remove outdated and pointless 'Last reviewed' lines from docs
They are very confusing for people, and more often than not
also just not very accurate. Seeing 'last reviewed: 2005' in
your docs is not very confidence-inspiring. Let's just remove
those comments.
2014-04-26 21:21:51 +01:00
Tim-Philipp Müller
a4f2720be4 clock: simplify internal gst_clock_return_get_name() helper 2013-08-14 21:41:23 +01:00
Wim Taymans
d73b4edc99 clock: debug the clock return values 2013-07-26 12:20:05 +02:00
Tim-Philipp Müller
666c8c11c6 Fix FSF address
https://bugzilla.gnome.org/show_bug.cgi?id=687520
2012-11-03 20:44:48 +00:00
Edward Hervey
d3ffa82639 Remove 0.10-related documentation and "Since" markers 2012-07-10 12:03:27 +02:00
Wim Taymans
10f3adfd9d clock: make abstract
Make the GstClock type abstract.
Fix a horrible hack in the clock unit test.
2012-07-09 16:28:22 +02:00
Tim-Philipp Müller
c0c79188ca bus, clock: make sure these never have a floating ref
Clear the initial floating ref in the init function for
busses and clocks. These objects can be set on multiple
elements, so there's no clear parent-child relationship
here. Ideally we'd just not make them derive from
GInitiallyUnowned at all, but since we want to keep
using GstObject features for debugging, we'll just do
it like this.

This should also fix some problems with bindings, which
seem to get confused when they get floating refs from
non-constructor functions (or functions annotated to
have a 'transfer full' return type). This works now:

from gi.repository import GObject, Gst

GObject.threads_init()
Gst.init(None)

pipeline=Gst.Pipeline()
bus = pipeline.get_bus()
pipeline.set_state(Gst.State.NULL)
del pipeline;

https://bugzilla.gnome.org/show_bug.cgi?id=679286
https://bugzilla.gnome.org/show_bug.cgi?id=657202
2012-07-09 13:09:45 +01:00
Wim Taymans
d7d5306009 clock: remove _full version
Rename gst_clock_id_wait_async_full() to gst_clock_id_wait_async()
and remove the old gst_clock_id_wait_async() version.
2012-06-20 09:58:14 +02:00
Evan Nemerson
6c6bb0e217 introspection: assorted introspection and documentation fixes
These changes are to clean up syntax issues such as missing colons,
missing spaces, etc., and minor issues such as argument names in
headers not matching the implementation and/or documentation.
2012-06-18 13:11:40 +02:00
Wim Taymans
ce67b0e539 docs: update more documentation 2012-03-29 13:34:50 +02:00
Wim Taymans
9aa9751938 review some docs 2012-03-28 18:12:23 +02:00
Wim Taymans
eb84682e92 clock: make more stuff private
Expose methods to get and set the timeout because subclasses uses this.
2012-02-27 09:11:36 +01:00
Wim Taymans
7cb4828e93 clock: remove method declaration too 2012-02-08 09:30:02 +01:00
Stefan Sauer
110e7c2d3c clock: remove unimplemented stats property while we can 2012-02-07 13:41:20 +01:00
Wim Taymans
f5886a17f5 remove TRACE_NAME from headers 2012-02-02 15:52:19 +01:00
Wim Taymans
321ee10bd6 trace: rework alloc tracing
Remove trace, we use debug log for that
Make alloc trace simpler, removing some methods.
Activate alloc trace with a GST_TRACE=3 environment variable.
Dump leaked objects atexit.
Provide an offset in the object where the GType can be found so that more
verbose info can be given for objects.
Remove -T option from gst-launch because tracing is now triggered with the
environment variable.
2012-01-27 17:54:43 +01:00
Wim Taymans
252327f87a Update for new gthread API 2012-01-19 09:27:04 +01:00
Tim-Philipp Müller
a1daf846f3 Merge remote-tracking branch 'origin/master' into 0.11
Conflicts:
	gst/gstobject.h
	libs/gst/check/gstcheck.h
	libs/gst/controller/gstcontroller.c
	plugins/elements/gstidentity.c
	tools/gst-xmlinspect.c
2011-12-04 15:38:09 +00:00
Tim-Philipp Müller
5889260d5a Work around deprecated thread API in glib master
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.
2011-12-04 13:35:38 +00:00
Wim Taymans
4a7c3b929a .h: fix header files
Ensure correct indentation and :retab.
Make sure all structures have padding
Fix up some old ABI additions.
2011-11-11 17:30:03 +01:00
Wim Taymans
8e4f006c57 fix more parent_class 2011-10-16 14:45:03 +02:00
Wim Taymans
5dd9ab1cab cleanups
Fix padding, remove deprecated symbols.
2011-02-22 16:04:12 +01:00
Wim Taymans
aa32e1fab1 clock: fix parameter docs 2011-02-10 15:21:45 +01:00
Sebastian Dröge
8163e51bad clock: API: Add function to re-init periodic GstClockIDs 2011-01-24 19:15:29 +01:00
Mart Raudsepp
3a0a6b477a docs: add missing "Since: 0.10.32" markers for GstClock
Since tags were missing for gst_clock_single_shot_id_reinit()
and GST_CLOCK_DONE.
2011-01-24 14:38:48 +00:00