Commit graph

425 commits

Author SHA1 Message Date
Vincent Penquerc'h
654a04f90c gstrtpssrcdemux: fix element leak 2012-01-12 18:23:42 +00:00
Tim-Philipp Müller
b8b8454bcb Suppress deprecation warnings in selected files, for g_static_rec_mutex_* mostly
GStaticRecMutex is part of our API/ABI, not much we can do here
in 0.10 for most of these.
2011-12-12 09:46:27 +00:00
Tim-Philipp Müller
330d984288 Use g_thread_try_new() instead of g_thread_crate() with newer glib versions 2011-12-12 09:46:27 +00:00
Tim-Philipp Müller
66f6e12888 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'll change these
over to the new API once we depend on glib >= 2.32.
2011-12-12 09:46:27 +00:00
Vincent Penquerc'h
c0e101e93f various: fix pad template leaks
https://bugzilla.gnome.org/show_bug.cgi?id=662664
2011-11-28 13:30:27 +00:00
Olivier Crête
1169bb05af gstrtpsession: Add special mode to use FIR as repair as Google does
https://bugzilla.gnome.org/show_bug.cgi?id=658419
2011-11-14 12:26:27 +01:00
Olivier Crête
79a9564c68 rtpsession: Send FIR requests in response to key unit requests with all-headers=TRUE
https://bugzilla.gnome.org/show_bug.cgi?id=658419
2011-11-14 12:26:27 +01:00
Olivier Crête
12a6b9613b rtpsession: Put the PLI requests in each RTPSource
Also refactor a bit and put all the keyframe request code in one
place inside rtpsession.c

https://bugzilla.gnome.org/show_bug.cgi?id=658419
2011-11-14 12:26:27 +01:00
Olivier Crête
59c028a4ce rtpsession: Hack to FIR because Google doesn't set the sender ssrc correctly
https://bugzilla.gnome.org/show_bug.cgi?id=658419
2011-11-14 12:26:27 +01:00
Olivier Crête
0ad78db0a3 rtpsession: Process received Full Intra Requests
Process FIR requests according to RFC 5104

https://bugzilla.gnome.org/show_bug.cgi?id=658419
2011-11-14 12:26:27 +01:00
Ha Nguyen
931020158e rtpbin: Fix a leaked clock for each buffering message
Fixes bug #659237.
2011-09-19 14:05:26 +02:00
Mark Nauwelaerts
e2179cbb74 rtpsession: avoid source premature timing out
Use slightly adjusted sender interval to determine sender timeout rather than
our own sender side interval (which may have been forced small).
2011-09-19 11:56:44 +02:00
Mark Nauwelaerts
f65d4c8300 rtpsession: avoid timing out source too quickly
... following a PAUSE/PLAY cycle, particularly applicable when operating
with a short RTCP interval (possibly forced so server-side).
2011-09-19 11:56:44 +02:00
Mark Nauwelaerts
77ebd33991 rtpjitterbuffer/rtpbin: relax dropping rtcp packets
... to at least having it trigger a/v synchronization, possibly without
using provided values which are still not considered sane
(as previously dropped).
2011-09-19 11:56:44 +02:00
Mark Nauwelaerts
adfe7d0467 rtpjitterbuffer: some more reset when clearing pt map
... which in particular caters for some more reset following a possible
rtsp PLAY.
2011-09-19 11:56:44 +02:00
Mark Nauwelaerts
915db26029 rtpjitterbuffer: only reset skew on gap if input ts available 2011-09-19 11:56:44 +02:00
Mark Nauwelaerts
1e17e10f75 rtpjitterbuffer: check some more for possible rtp timestamp discontinuity
... when operating in non slave mode, and reset if detected.
This should avoid some (large) bogus outgoing timestamp due to jumps
in rtp time, as result of PAUSE/PLAY or seek or ...
2011-09-19 11:56:40 +02:00
Mark Nauwelaerts
9c95072048 rtpbin: alternative inter-stream syncing methods
... at least if not syncing to NPT time:
* either sync using RTCP SR data (as currently)
* only perform the above once using initial RTCP SR packets
* discard RTCP and sync by equating provided stream's clock-base rtptime,
  as provided by jitterbuffer (typically obtained from RTP-Info in RTSP).
2011-09-19 11:52:03 +02:00
Mark Nauwelaerts
4b7301e4d1 rtpjitterbuffer: also provide clock-base to sync signal 2011-09-19 11:52:00 +02:00
Mark Nauwelaerts
f29c253934 rtpbin: allow configurable rtcp stream syncing interval
... rather than necessarily syncing at each RTCP SR.
2011-09-19 11:51:57 +02:00
Mark Nauwelaerts
afd26f0078 rtpsession: trigger reconsideration if rtcp interval set 2011-09-19 11:51:50 +02:00
Olivier Crête
b2e8362767 rtpsession: Initialise the last_keyframe_request variable 2011-09-02 19:24:46 -04:00
Mark Nauwelaerts
c03648c8bb rtpsession: properly init rtcp_min_interval 2011-07-29 12:08:42 +02:00
Mark Nauwelaerts
3a98f6f0fd rtpssrcdemux: keep a ref on the src pad while using it
Prevent a possible race if clear_ssrc() is called between getting the pad and
doing the push.

Based on patch by <olivier.crete@collabora.com>

https://bugzilla.gnome.org/show_bug.cgi?id=650916
2011-07-28 14:51:01 +02:00
Olivier Crête
c7b9b98648 rtpssrcdemux: Make the pads lock recursive and hold it across the signal emit
We need to keep the lock held because we don't want a push before the "new-ssrc-pad"
handler has completed. But we may want to push an event from inside that handler, hence
the recursive mutex.

https://bugzilla.gnome.org/show_bug.cgi?id=650916
2011-07-28 14:50:59 +02:00
Olivier Crête
e26b5391c2 rtpssrcdemux: Use PADs lock
https://bugzilla.gnome.org/show_bug.cgi?id=650916
2011-07-28 14:50:57 +02:00
Olivier Crête
6095d2a3f0 rtpsession: Always send application requested feedback in immediate mode
Send as many application requested feedback messages in immediate mode, even if they
have already been sent.

https://bugzilla.gnome.org/show_bug.cgi?id=654583
2011-07-25 17:20:59 +02:00
Olivier Crête
354faabda0 rtpsession: Don't let the computed RTP bandwidth fall too low
If it falls too low, the computed RTCP bandwidth will be near zero and
the RTCP thread will be stopped.

https://bugzilla.gnome.org/show_bug.cgi?id=654583
2011-07-25 16:19:00 +02:00
Olivier Crête
4d48109f9d rtpsession: Wait longer to timeout SSRC collision
Using the current RTCP interval to timeout SSRC collision can lead to
collisions being timed out immediately if a BYE packet is sent because
it is sent immediately, so the interval is 0. This is not what we
want. So just set a static 10 times the default RTCP interval, it
should be enough

https://bugzilla.gnome.org/show_bug.cgi?id=648642
2011-07-25 16:18:58 +02:00
Miguel Angel Cabrera Moya
977a5eee7a rtpjitterbuffer: return correct type when assertion fails 2011-06-24 11:59:01 +02:00
Olivier Crête
581a30d892 rtpsession: The signal has 5 arguments, not 4 2011-06-20 16:47:36 -04:00
Sebastian Dröge
b694bfeca3 ssrcdemux: Fix uninitialized variable compiler warning for (pre-) releases too 2011-05-17 10:47:32 +02:00
Sebastian Dröge
0f05d3e5a5 rtpssrcdemux: Fix uninitialized variable compiler warning 2011-05-17 09:24:08 +02:00
Olivier Crête
b6bfc512e8 ssrcdemux: Implement iterate internal links for sink pads
https://bugzilla.gnome.org/show_bug.cgi?id=649617
2011-05-17 09:22:29 +02:00
Olivier Crête
23b6c8febc rtpssrcdemux: iterate pad function is only valid for src pads
The iterate function is only used for src pads, so mark it as such and remove
dead code.

https://bugzilla.gnome.org/show_bug.cgi?id=649617
2011-05-17 09:22:25 +02:00
Olivier Crête
1bf94a92b0 rtpssrcdemux: Release lock before emitting signal
If the lock is not released before emitting a signal, it may cause a deadlock
if any other function in the element is called.

Also removed an unused timestamp parameter

https://bugzilla.gnome.org/show_bug.cgi?id=649617
2011-05-17 09:22:20 +02:00
Olivier Crête
42531337f5 rtpsession: Remove incomplete support for RTCP FIR
Remove bits that were meant to suppport RTCP FIR

https://bugzilla.gnome.org/show_bug.cgi?id=648160
2011-04-20 07:50:43 +01:00
Robert Swain
5b18c652fb rtp, rtpmanager: Address unused but set variables
GCC 4.6.x spits warnings about variables that are unused but set. Such
variables have been removed where trivial but with comments left behind
for informational purposes in some cases.

gst_rtp_session_chain_recv_rtcp () was changed in commit 490113d4
to always return GST_FLOW_OK instead of the return value of
rtp_session_process_rtcp (), so we'll keep it that way.
2011-04-16 12:49:16 +01:00
Olivier Crête
9d9257916b rtpsession: Use existing functions to parse RTCP FB packets
Use existing functions to get the FCI from FB packets.

https://bugzilla.gnome.org/show_bug.cgi?id=622553
2011-04-15 12:48:04 +01:00
Olivier Crête
5ccd964d86 rtpsession: marshal GstBuffer as a MiniObject instead of a pointer
https://bugzilla.gnome.org/show_bug.cgi?id=622553
2011-04-15 12:47:40 +01:00
Pascal Buhler
0d2d52856f rtpssrcdemux: Unknown SSRC is not fatal
https://bugzilla.gnome.org/show_bug.cgi?id=646966
2011-04-11 17:37:58 -04:00
Pascal Buhler
58ef84846e rtpsession: Number of active sources should be updated whenever the status of the source changes to active
Forward-ported by Olivier Crête

https://bugzilla.gnome.org/show_bug.cgi?id=646965
2011-04-11 17:37:36 -04:00
Havard Graff
53c88ae33e rtpmanager: ignore a BYE if it is sent with our internal SSRC
https://bugzilla.gnome.org/show_bug.cgi?id=646964
2011-04-11 17:34:12 -04:00
Thibault Saunier
b541208b77 android: Make it ready for androgenizer
Remove the android/ top dir
Fixe the Makefile.am to be androgenized

To build gstreamer for android we are now using androgenizer which generates the needed Android.mk files.
Androgenizer can be found here: http://git.collabora.co.uk/?p=user/derek/androgenizer.git
2011-04-11 01:20:11 +02:00
Havard Graff
93f022d6ab rtpsession: fix wrongly applied patch
Obviously recv_rtp_sink does not have much to do with send_rtcp_src...
See commit 046ff170.

https://bugzilla.gnome.org/show_bug.cgi?id=647263
2011-04-09 12:32:37 +01:00
Havard Graff
e71a908d96 jitterbuffer: Make src_query MT-safe
It is possible that the element might be going down while the event arrives
2011-04-08 15:23:05 +02:00
Sebastian Dröge
4c36ca30b2 jitterbuffer: Unref event if the parent element disappeared 2011-04-08 15:22:19 +02:00
Havard Graff
342686bb02 jitterbuffer: Make upstream events MT-safe 2011-04-08 15:21:46 +02:00
Sebastian Dröge
31af4fe33e rtp: Unref events if the parent element disappeared 2011-04-08 15:20:51 +02:00
Ole André Vadla Ravnås
046f170d6a rtpmanager: fix pad callbacks so they handle when parent goes away
1) We need to lock and get a strong ref to the parent, if still there.
2) If it has gone away, we need to handle that gracefully.

This is necessary in order to safely modify a running pipeline. Has been
observed when a streaming thread is doing a buffer_alloc() while an
application thread sends an event on a pad further downstream, and from
within a pad probe (holding STREAM_LOCK) carries out the pipeline plumbing
while the streaming thread has its buffer_alloc() in progress.
2011-04-08 15:16:56 +02:00