Commit graph

167 commits

Author SHA1 Message Date
Nicolas Dufresne
ed5ac1c101 doc: Silence warning about unused gstntppacket section
This API is internal.
2015-06-13 14:40:43 -04:00
Nicolas Dufresne
d7b22481ae Revert "doc: Add GstNtpPacket to the doc"
This reverts commit c4eb876961.

Oops, this is not a public API
2015-06-13 14:37:48 -04:00
Nicolas Dufresne
c4eb876961 doc: Add GstNtpPacket to the doc 2015-06-13 14:21:41 -04:00
Tim-Philipp Müller
1164f91f1f libs: more doc scanner fixes
gstbasetransform.h:196: Warning: GstBase: "@submit_input_buffer" parameter unexpected at this location:
 * @submit_input_buffer: Function which accepts a new input buffer and pre-processes it.

gstnetcontrolmessagemeta.c:103: Warning: GstNet: gst_buffer_add_net_control_message_meta: unknown parameter 'message' in documentation comment, should be 'addr'
2015-06-13 15:06:11 +01:00
Sebastian Dröge
59d916a071 netclientclock: Use the new GST_CLOCK_FLAG_NEEDS_STARTUP_SYNC flag
https://bugzilla.gnome.org/show_bug.cgi?id=750574
2015-06-09 10:03:49 +02:00
Sebastian Dröge
558c0b97fc netclientclock: Make the clock a wrapper clock around an internal clock
The internal clock is only used for slaving against the remote clock, while
the user-facing GstClock can be additionally slaved to another clock if
desired. By default, if no master clock is set, this has exactly the same
behaviour as before. If a master clock is set (which was not allowed before),
the user-facing clock is reporting the remote clock as internal time and
slaves this to the master clock.

This also removes the weirdness that the internal time of the netclientclock
was always the system clock time, and not the remote clock time.

https://bugzilla.gnome.org/show_bug.cgi?id=750574
2015-06-09 10:03:49 +02:00
Sebastian Dröge
5ba9e9d9a8 ptp: Make sure to always initialize the variables we put into the statistics structure later
CID 1304676, 1304677, 1304678, 1304679.
2015-06-08 12:22:56 +02:00
Руслан Ижбулатов
499b8d841e ptp: Fix build on Windows, and in general the GI build when PTP support was not available
It's not going to work on Windows still, the helper process needs to be
ported.
2015-06-08 10:46:24 +02:00
Sebastian Dröge
41f8e03b85 netclientclock: The NTP poll interval is a signed int8, not unsigned 2015-06-07 10:52:33 +02:00
Sebastian Dröge
39a3d0d87d netclientclock: Add Since marker to the docs for gst_ntp_clock_new() 2015-06-07 09:32:39 +02:00
Sebastian Dröge
64773fee9c ptpclock: Use the current path delay for calculation the local/remote clock times
The mean might currently be changing, and the current path delay is the
closest we can get to the actual delay around the current SYNC message.
2015-06-07 08:59:23 +02:00
Sebastian Dröge
9e0f051990 netclientclock: Add some copyright stuff 2015-06-06 23:05:32 +02:00
Sebastian Dröge
202f463170 netclientclock: Add NTPv4 support
This uses all of the netclientclock code, except for the generation and
parsing of packets. Unfortunately some code duplication was necessary
because GstNetTimePacket is public API and couldn't be extended easily
to support NTPv4 packets without breaking API/ABI.
2015-06-06 23:00:44 +02:00
Sebastian Dröge
cc15f68b6a netclientclock: Preparation for NTPv4 support
We extend our calculations to work with local send time, remote receive time,
remote send time and local receive time. For the netclientclock protocol,
remote receive and send time are assumed to be the same value.

For the results, this modified calculation makes absolutely no difference
unless the two remote times are different.
2015-06-06 20:39:47 +02:00
Sebastian Dröge
8d6837edf5 netclientclock. Fix last commit
Apparently I failed at git add -i.
2015-06-06 19:01:06 +02:00
Sebastian Dröge
4c62d791aa netclientclock: Make gst_net_client_clock_new() a thing wrapper around g_object_new()
Bindings will like this, and also it fixes a FIXME comment.
2015-06-06 18:42:18 +02:00
Sebastian Dröge
179eebdf44 ptpclock: Use #define everywhere instead of G_N_ELEMENTS() 2015-06-06 14:34:39 +02:00
Sebastian Dröge
f7a7a569cd netclientclock: Filter RTTs based on the median of the last RTTs before considering them at all
This improves accuracy on wifi or similar networks, where the RTT can go very
high up for a single observation every now and then. Without filtering them
away completely, they would still still modify the average RTT, and thus all
clock estimations.
2015-06-06 14:31:16 +02:00
Sebastian Dröge
42b16c9e3c ptpclock: Use a system clock for the time observations instead of gst_util_get_timestamp()
They don't necessarily use the same underlying clocks (e.g. on Windows), or
might be configured to a different clock type (monotonic vs. real time clock).

We need the values a clean system clock returns, as those are the values used
by the internal clocks.
2015-06-06 14:19:21 +02:00
Sebastian Dröge
bf62a96549 ptpclock: Fix documentation a bit 2015-06-06 12:36:34 +02:00
Sebastian Dröge
d2da80e4e7 ptp: Fix debug output to print the difference instead of absolute values 2015-06-03 16:09:17 +02:00
Wim Taymans
e11568c5b8 ptpclock: fix compilation
Don't put code between declarations.
Fix use of uninitialized variables
2015-06-03 15:22:31 +02:00
Sebastian Dröge
ac2eb6e23e ptp: Add median based pre-filtering of delays
If the delay measurement is too far away from the median of the window of last
delay measurements, we discard it. This increases accuracy on wifi a lot.

https://bugzilla.gnome.org/show_bug.cgi?id=749391
2015-06-03 13:55:39 +02:00
Sebastian Dröge
02ba5b20c7 ptp: Add #define to only use SYNC messages for which we can send DELAY_REQ
https://bugzilla.gnome.org/show_bug.cgi?id=749391
2015-06-03 13:55:36 +02:00
Sebastian Dröge
5e1974fc08 ptp: Add #defines to enable/disable improvements for unreliable networks
We should do some more measurements with all these and check how much sense
they make for PTP. Also enabling them means not following IEEE1588-2008 by the
letter anymore.

https://bugzilla.gnome.org/show_bug.cgi?id=749391
2015-06-03 13:55:32 +02:00
Sebastian Dröge
8d77759834 ptp: Initial implementation of a PTP clock
GstPtpClock implements a PTP (IEEE1588:2008) ordinary clock in
slave-only mode, that allows a GStreamer pipeline to synchronize
to a PTP network clock in some specific domain.

The PTP subsystem can be initialized with gst_ptp_init(), which then
starts a helper process to do the actual communication via the PTP
ports. This is required as PTP listens on ports < 1024 and thus
requires special privileges. Once this helper process is started, the
main process will synchronize to all PTP domains that are detected on
the selected interfaces.

gst_ptp_clock_new() then allows to create a GstClock that provides the
PTP time from a master clock inside a specific PTP domain. This clock
will only return valid timestamps once the timestamps in the PTP domain
are known. To check this, the GstPtpClock::internal-clock property and
the related notify::clock signal can be used. Once the internal clock
is not NULL, the PTP domain's time is known. Alternatively you can wait
for this with gst_ptp_clock_wait_ready().

To gather statistics about the PTP clock synchronization,
gst_ptp_statistics_callback_add() can be used. This gives the
application the possibility to collect all kinds of statistics
from the clock synchronization.

https://bugzilla.gnome.org/show_bug.cgi?id=749391
2015-06-03 13:55:28 +02:00
Tim-Philipp Müller
d71f5c7d73 net: keep GCancellable fd around instead of re-creating it constantly
Just create the cancellable fd once and keep it around instead
of creating/closing it for every single packet. Since we spend
most time waiting for packets, an fd is alloced and in use pretty
much all the time anyway.
2015-05-19 14:34:04 +01:00
Tim-Philipp Müller
8e6561be91 Remove obsolete Android build cruft
This is not needed any longer.
2015-04-26 17:49:44 +01:00
Edward Hervey
e4d07e161d introspection: Don't use g-ir-scanner cache at compile time
It pollutes user directories and we don't need to cache it

https://bugzilla.gnome.org/show_bug.cgi?id=747095
2015-03-31 11:16:09 +02:00
William Manley
fc765592a1 meta: Add GstNetControlMessageMeta
GstNetAddress can be used to store ancillary data which was received with
or is to be sent alongside the buffer data.  When used with socket sinks
and sources which understand this meta it allows sending and receiving
ancillary data such as unix credentials (See `GUnixCredentialsMessage`)
and Unix file descriptions (See `GUnixFDMessage`).

This will be useful for implementing protocols which use file-descriptor
passing in payloaders/depayloaders without having to re-implement all the
socket handling code already present in elements such as multisocketsink,
etc.  This, in turn, will be useful for implementing zero-copy video IPC.

This meta uses the platform independent `GSocketControlMessage` API
provided by GLib as a part of GIO.  As a result this new meta does not
require any new dependencies or any conditional compliation for
portablility, although it is unlikely to do anything useful on non-UNIX
platforms.
2015-03-14 13:54:51 +01:00
Jan Schmidt
f0e3b7f904 netclock: Don't update the clock when it desynch
Add a hold off when the clock calibration suddenly loses synch,
as it may be a glitch, but also make sure we update if it stays
desynched for more than a few seconds
2015-02-07 04:41:52 +11:00
Jan Schmidt
824b821caa netclock: Make the RTT average ignore large values more forcefully.
Smooth larger RTTs a little harder, so excessively large values
perturb the average a bit less, and therefore get filtered out
more strongly
2015-02-07 04:38:42 +11:00
Jan Schmidt
aeca7eb480 netclock: Fix docs typo. Clock bus messages are GST_MESSAGE_ELEMENT 2015-01-21 22:44:59 +11:00
Jan Schmidt
27997cb8a3 netclock: Implement rate limits for polling and fix up skew limits
Add the minimum-update-interval property to the clock, with a default
of 50ms and don't send polling requests faster than that. That helps to
ensure we spread the initial observations out a little - startup takes
a little longer, but tracking is more stable.

Move the discont skew limiting code inside an if statement, so that
it's only done when the linear regression succeeds and the clock
parameters might actually change.
2015-01-21 22:27:18 +11:00
Jan Schmidt
1c9c9847fd netclock: Implement sending statistic bus messages and discont limits
Allow setting a GstBus on the network clock client
via a new 'bus' object property. If a bus is set, the
clock will output an element message containing statistics
about new clock observations and the clock correlation.

When the local clock is synchronised with the remote, limit the
maximum jump in the clock at any point to be one average RTT to
the server. Also, publish in the bus message whether we are
synched with the remote or not.
2015-01-21 22:27:18 +11:00
Sebastian Rasmussen
863ba1a598 docs: Fix documentation typos and inconsistencies
* GstGlobalDeviceMonitor was renamed to GstDeviceMonitor
 * Expand GST_MESSAGE_DEVICE to the full enum value names
 * Correct the incorrect references to the GstDeviceProvider interfaces
 * Describe caps arguments for gstcheck interface
 * Add missing docs for GstNetAddressMeta and its add function
 * Add docs for toc helper macros
 * Avoid refering to GstValueList type as done elsewhere

Fixes https://bugzilla.gnome.org/show_bug.cgi?id=732786
2014-07-09 09:22:11 +02:00
Edward Hervey
bf9018b77c nettimeprovider: Use non-freed variable
address is only used temporarily. Use the proper variable instead.

CID #1212189
2014-05-09 14:46:59 +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
Antoine Jacoutot
3e24bbefba libs: g-ir-scanner: do not hardcode libtool path
https://bugzilla.gnome.org/show_bug.cgi?id=726571
2014-04-17 09:26:20 +01:00
Tim-Philipp Müller
2ddbe8df49 nettimeprovider: remove unnecessary NULL check
Error is never NULL when we break out of the loop.

COVERITY CID 1037151
2013-12-30 16:10:08 +00:00
Tim-Philipp Müller
0735c052e1 Revert "nettimeprovider: Remove dead code"
This reverts commit 9649cd4ca1.

This is not right, and it's also not what coverity
is complaining about.
2013-12-30 16:07:08 +00:00
Edward Hervey
9649cd4ca1 nettimeprovider: Remove dead code
err is always NULL by the point we reach this line

COVERITY CID 1037151
2013-12-30 16:51:01 +01:00
Sebastian Rasmussen
53ae1b2c9c docs: Fix typos in function/object descriptions
Fixes https://bugzilla.gnome.org/show_bug.cgi?id=720029
2013-12-07 17:11:12 +00:00
Jan Schmidt
73a58cb0a2 netclock: Fix docstring for round-trip-limit and uninit access warning.
Fix a typo in a doc string - the property is round-trip-limit, not
roundtrip-limit.

Remove a bogus GST_WARNING that can print an uninitialised variable
and is redundant anyway.
2013-11-27 18:32:22 +11:00
Carlos Rafael Giani
1ce704d45d netclock: Add round-trip-limit parameter
Sometimes, packets might take a very long time to return. Such packets
usually are way too late and destabilize the regression with their
obsolete data. On Wi-Fi, round-trips of over 7 seconds have been observed.

If the limit is set to a nonzero value, packets with a round-trip period
larger than the limit are ignored.

Signed-off-by: Carlos Rafael Giani <dv@pseudoterminal.org>

https://bugzilla.gnome.org/show_bug.cgi?id=712385
2013-11-27 18:15:20 +11:00
Jan Schmidt
75aa78ca70 netclock: Fix C99 comment 2013-11-26 02:43:54 +11:00
Jan Schmidt
5b2df52c6a netclock: Implement rolling-average filter on observations.
Keep a rolling average of the round trip time for network clock
observations, favouring shorter round trips as being more accurate.

Don't pass any clock observation to the clock slaving if it has a
round-trip time greater than 2 times the average.

Actual shifts in the network topology will be noticed after some
time, as the rolling average incorporates the new round trip times.
2013-11-26 02:20:14 +11:00
Philip Withnall
3c8424d9ae net: Constify a parameter to gst_net_client_clock_new()
Even though this parameter is not used, it should be const to fit in with the
coding standards for other similar parameters. Client code already passes in
const strings under the expectation that they won’t be modified.

https://bugzilla.gnome.org/show_bug.cgi?id=710442
2013-10-18 09:59:25 +01:00
Tim-Philipp Müller
98c1342614 timeprovider: g-i: allow None as address for gst_net_time_provider_new() 2013-07-16 14:48:18 +01:00
Sebastian Dröge
d420686764 netclock: Add support for IPv6 2013-04-24 15:59:45 +02:00