Commit graph

447 commits

Author SHA1 Message Date
Jan Schmidt
43209d0ee6 dashdemux2: Preserve current representation on live manifest updates
When updating a manifest during live playback, preserve the current
representation for each stream.

During update_fragment_info, if the current representation changed
because it couldn't be matched, trigger a caps change and new
header download.

This reverts commit e0e1db212f
and reapplies "dashdemux: Fix issue when manifest update sets slow start
without passing necessary header & caps changes downstream" with
changes.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2920>
2022-09-05 16:07:00 +00:00
Sebastian Dröge
cc454f0fc3 rtpjitterbuffer: Add test for crash caused by removing timers twice
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2973>
2022-09-03 09:26:24 +00:00
Sebastian Dröge
648b8f3362 rtpjitterbuffer: Make it more explicit that update_rtx_timers() takes ownership of the passed in timer
It is not valid anymore afterwards and must not be used, otherwise an
already freed pointer might be used.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2973>
2022-09-03 09:26:24 +00:00
Sebastian Dröge
e66f5e2423 rtpjitterbuffer: Don't shadow variable
While this didn't cause any problems in this context it is simply
confusing.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2973>
2022-09-03 09:26:24 +00:00
Sebastian Dröge
0b19c457ca rtpjitterbuffer: Change RTX timer availability checks to assertions
It's impossible to end up in the corresponding code without a timer for
RTX packets because otherwise it would be an unsolicited RTX packet and
we would've already returned early.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2973>
2022-09-03 09:26:24 +00:00
Sebastian Dröge
2ca849499e rtpjitterbuffer: Only unschedule timers for late packets if they're not RTX packets and only once
Timers for RTX packets are dealt with later in update_rtx_timers(), and
timers for non-RTX packets would potentially also be unscheduled a
second time from there so avoid that.

Also don't shadow the timer variable from the outer scope but instead
make use of it directly.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2973>
2022-09-03 09:26:24 +00:00
Patricia Muscalu
3c9e4f4886 rtph265: keep delta unit flag
Without this patch all buffers that pass the payloader
are marked as non-delta-unit buffers.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2969>
2022-09-02 08:56:13 +00:00
Thibault Saunier
6a4425e46a meson: Call pkgconfig.generate in the loop where we declare plugins dependencies
Removing some copy pasted code

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2970>
2022-09-01 21:17:35 +00:00
Thibault Saunier
bc9c1e3956 meson: Namespace the plugins_doc_dep/libraries variables
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2970>
2022-09-01 21:17:35 +00:00
Thibault Saunier
b5e90fe579 meson: Rename plugins list and make them "dependency" objects
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2970>
2022-09-01 21:17:35 +00:00
Raul Tambre
e1d3612321 rtpjitterbuffer: remove lost timer for out of order packets
When receiving old packets remove the running lost timer if present.
This fixes incorrect reporting of a lost packet even if it arrived in time.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2922>
2022-09-01 09:01:31 +00:00
Sebastian Dröge
cbc6761199 rtpvp8depay: If configured to wait for keyframes after packet loss, also do that if incomplete frames are detected
This can happen if the data inside the packets is incomplete without the
seqnums being discontinuous because of ULPFEC being used.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2947>
2022-08-31 08:58:03 +00:00
Stéphane Cerveau
6b1ffadb0c docs: disable in static build
Following gst-plugins-base, disable docs if static_build
in:
 - gstreamer
 - gst-plugins-good
 - gst-plugins-ugly
 - gst-libav
 - gstreamer-vaapi

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2529>
2022-08-30 14:46:28 +00:00
Jianhui Dai
625e74100d v4l2allocator: Fix invalid imported dmabuf fd
Fix a typo that set userptr to dmabuf fd. It leads to failure of
dmabuf-import io-mode.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2935>
2022-08-24 09:59:14 +00:00
Mathieu Duponchelle
8756f523d1 playback: add onvif metadata caps to raw caps
+ remove encoding from x-onvif-metadata caps output by qtdemux

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2889>
2022-08-24 12:21:18 +03:00
zhiyuan.liu
ffebd52e46 isoff: Fix earliest pts field parse issue
earliest pts will be covered by first_offset field on version 0 case.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2927>
2022-08-23 10:59:56 +00:00
Jan Alexander Steffens (heftig)
5ae3c9318d adaptivedemux2: Fix download helper with libsoup 3.0.x
libsoup 3.0.x dispatches using a single source attached when the session
is created, so we need to create the session with the same context that
our download thread is later using.

2.74 or 3.1 will dispatch a response using the context which sent the
request. However, for any context other than the one that created the
session, this will also create and destroy sources, so there's still
some slight performance benefit.

Fixes: https://gitlab.freedesktop.org/gstreamer/gstreamer/-/issues/1384
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2913>
2022-08-18 20:16:18 +00:00
Jan Schmidt
ba8f14e922 adaptivedemux2: Implement send_event() vfunc
Handle select-streams and seek events in an element
level send_event() vfunc, so they can be received
before any source pads are created.

This allows preferred streams to be selected before
segment downloading starts.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2912>
2022-08-18 15:30:05 +01:00
Jan Schmidt
4a6c2e6720 splitmuxsrc: Stop pad task before cleanup
When stopping the element, make sure the pad task
is stopped before destroying the part readers.

Closes a race where the pad task might access
a freed pointer.

Also add a guard against this sort of thing
by holding a ref to the reader in the pad loop.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2901>
2022-08-17 09:42:50 +10:00
Jan Schmidt
c2fa0b50ce qtdemux: Avoid crash on reconfiguring.
When reconfiguring a stream that never created
an output pad, don't access a NULL GstPad pointer.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2869>
2022-08-16 19:01:28 +00:00
Sebastian Dröge
a3037eb453 qtdemux: Set parsed=true on ONVIF Timed Metadata caps
Inside MP4 the metadata must be properly parsed into frames and in
order.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2897>
2022-08-16 18:11:53 +00:00
Nicolas Dufresne
d126f65485 v4l2bufferpool: Fix debug trace
The tracing of the buffer pointer was done on the secondary pointer, which
does not match with other traces of the same buffer. This made the trace
confusing and less useful.

Fixes #1379

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2893>
2022-08-16 14:33:24 +00:00
Jan Schmidt
f4f1819416 adaptivedemux2: Fix for period switching in live streams
When playing live, it's possible that one stream reaches
the end of the available playback window and goes to sleep
waiting for a manifest update, and the manifest update
introduces a new period. In that case, the sleeping
stream needs to wake up and go 'properly' EOS before we
can advance the input to the new period.

Accordingly, make sure that a stream's last_ret value
is not marked as EOS if it's just sleeping waiting for a live
manifest update.

Also fix the output loop to go back and re-check if it's
time to switch to the next period after dequeuing and
discarding an EOS event.

https://livesim.dashif.org/livesim/periods_20/testpic_2s/Manifest.mpd

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2895>
2022-08-16 12:39:58 +00:00
Hosang Lee
3cd79c4262 mssdemux2: Use gsturi structure to form fragment urls
Utilize gsturi to form fragment url paths.
A token query may contain the string "manifest" and this would lead
to improper url creations.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2859>
2022-08-16 17:57:19 +09:00
Sebastian Dröge
8e77c8155c rtspsrc: Consider the actual control base URI also in case the connection URI contains a query string
That is, get rid of unnecessary and wrong special-casing.

This could always use gst_rtsp_url_get_request_uri_with_control() but as
we only have the control base URI as string it is easier to just call
gst_uri_join_strings().

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2868>
2022-08-12 18:52:29 +00:00
Sebastian Dröge
b0533d1ea0 qtdemux: Add reference timestamp meta with UTC times based on the ONVIF Export File Format CorrectStartTime box to outgoing buffers
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2525>
2022-08-12 16:13:50 +00:00
Jan Schmidt
025e446470 hlsdemux2: Fix reference leak of variant stream
When switching back to the previous variant stream
in gst_hls_demux_change_playlist(), fix a couple of
paths that would leak a reference to the previous
variant.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2873>
2022-08-12 13:52:59 +00:00
Jan Schmidt
9db19a3b06 hlsdemux2: Don't leak the datetime in time map structs
Add a function to clean up GstHLSTimeMap structs
and free the ref on the optional associated GDateTime

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2873>
2022-08-12 13:52:59 +00:00
Jan Schmidt
a242428532 hlsdemux2: Fix typefind leak and invalid memory access
When typefinding aggregates incoming data to a pending
typefind buffer and then succeeds in typefinding, it
leaks the aggregated buffer, and leaves the caller
accessing an unreffed buffer.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2873>
2022-08-12 13:52:59 +00:00
Jan Schmidt
a6eb725413 hlsdemux2: m3u8: Fix memory leaks on parsing
Fix memory leaks when parsing of an m3u8 file is
incomplete, with EXTINF or EXT-X-PROGRAM-DATE-TIME
directives, but no segment url.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2873>
2022-08-12 13:52:59 +00:00
Jan Schmidt
0ff317f107 hlsdemux2: m3u8: Fix memory leak
Clear the GValue holding intermediate GstStructure field
data.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2873>
2022-08-12 13:52:59 +00:00
Jan Schmidt
fd64e8c7ed adaptivedemux2: Fix uninitialised memory usage in debug
Fix printing uninitialised memory by clearing the
GstAdaptiveDemuxClock structure when allocating.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2873>
2022-08-12 13:52:59 +00:00
Nirbheek Chauhan
d8c4ebccab rtpst2022-1-fecenc: Drain column packets on EOS
Otherwise we won't send the protection packets for the last few
packets when a stream ends.

Also send EOS on the FEC src row pad immediately, and on the FEC src
column pad after draining is complete. This makes it so that the FEC
src pads on rtpbin behave the same way as the RTCP src pads on rtpbin
when EOS is received on the send_rtp_sink pad.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2863>
2022-08-12 12:59:19 +00:00
Edward Hervey
63dcee34fb qtdemux: Don't use invalid values from failed trex parsing
If parsing the fragment default values (`trex` atom) failed, don't try to
compute a bogus sample_description_id value.

Fixes #1369

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2860>
2022-08-11 08:50:34 +02:00
Piotr Brzeziński
c883a9f54b videoflip: Add support for 10/12bit planar formats
Implements support for I420, I422 and Y444 in 10/12 bit LE/BE variants.
I422 is handled separately from the rest, as it needs to consider
the endianness of the current format during most transforms.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2788>
2022-08-10 10:52:27 +00:00
Haihua Hu
82025897c4 alpha: fix stride issue when out buffer has padding on right
if outbuf has padding on right, need jump to next line use stride,
otherwise downstream element will show a wrong picture when use the
same stride

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2842>
2022-08-09 13:04:11 +08:00
Jan Schmidt
edf84a7119 hlsdemux2: Requeue header buffer when restarting fragment
When returning GST_ADAPTIVE_DEMUX_FLOW_RESTART_FRAGMENT
for the first segment data, we might need to requeue the
header.

This was leading to occasional prerolling stalls on
HLS live streams with renditions.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2849>
2022-08-08 15:52:22 +00:00
Jan Schmidt
91e43048cf hlsdemux2: Fix buffer leak when resynching
Unref the buffer in gst_hls_demux_handle_buffer() when
returning GST_ADAPTIVE_DEMUX_FLOW_LOST_SYNC

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2849>
2022-08-08 15:52:22 +00:00
Jan Schmidt
b7e662f400 adaptivedemux2: Fix a small race on shutdown
Make sure gst_adaptive_demux_loop_cancel_call()
never tries to operate on an invalidated main context. Make
sure to clear the main context pointer while holding the lock,
and to check it in gst_adaptive_demux_loop_cancel_call()

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2847>
2022-08-05 23:36:49 +01:00
Jan Schmidt
d60c2f46e2 adaptivedemux2-stream: Silence a compiler warning
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2839>
2022-08-04 14:49:49 +01:00
Jan Schmidt
1060f72f90 adaptivedemux2: Move internal FLOW_SWITCH return value.
Move the internal-only FLOW_SWITCH custom return value
to GST_FLOW_CUSTOM_SUCCESS+2 to avoid collision with
GST_ADAPTIVE_DEMUX_FLOW_LOST_SYNC

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2839>
2022-08-04 14:49:33 +01:00
Edward Hervey
0857f5af5f adaptivedemux2: Modify custom sync loss flow return
Make it a custom sucess and not an error

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2839>
2022-08-04 14:49:33 +01:00
Edward Hervey
6934362817 hlsdemux2: Always check DSN if required
We don't want to consider the candidate as being before the playlist if the DSN
don't match

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2839>
2022-08-04 14:49:27 +01:00
Edward Hervey
5d0b112c0c adaptivedemux2/hlsdemux2: Handle loss of sync when dowloading.
Media playlist updates and fragment downloads happen in an interleaved
fashion. When a media playlist update fails *while* a segment is being
downloaded, this means we lost synchronization.

Properly propagate and handle this

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2839>
2022-08-04 14:49:27 +01:00
Edward Hervey
6a9bdceb5e hlsdemux2: Fix initial playlist setup.
There is now only a single case where we setup the initial playlist to 0, which
is for the very first variant stream.

Rendition streams will have the initial playlist "synchronized" against the
variant stream media playlist.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2839>
2022-08-04 14:49:27 +01:00
Edward Hervey
9ca1dcd563 hlsdemux2: Handle loss of synchronization in live
Loss of synchronization happens when the updated media playlist has no
relationship to the previous ones. This could happen because of network issues,
server issues, etc...

When this happens, we take no chance and "reset" ourselves so that we can "seek
back to live" against the new updated playlists.

Since this happens at the "media playlist update" level, make sure the custom
flow return is propagated up.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2839>
2022-08-04 14:49:27 +01:00
Edward Hervey
abe5a06421 adaptivedemux2: Handle synchronously to lost sync
We are already in the main scheduler thread, therefore we can do the "seek back
to live" directly. This also avoids other pending actions to take place.

Also handle the loss of sync when doing manifest updates.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2839>
2022-08-04 14:49:27 +01:00
Edward Hervey
1aab9db803 hlsdemux2: Prune time maps when possible
Add a new method to prune unused time mappings (i.e. which aren't used by any
current media playlist).

Do that when doing flushing seeks. Could be used in other places later too.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2839>
2022-08-04 14:49:21 +01:00
Edward Hervey
c20ad629c7 hlsdemux2: Allow DSN mismatches when re-syncing playlists
Some providers provide completely incompatible DSN across bitrates/renditions,
but do keep MSN consistent.

If we fail to synchronize playlist with DSN, retry without the DSN taken into
account.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2839>
2022-08-04 14:49:12 +01:00
Jan Schmidt
da65a7b0ba adaptivedemux2: Fixes for period switching in the output loop
Close some race conditions in switching to the next period,
by ensuring the tracks are completely drained first and by
not outputting EOS events to the output source pad
if there is another period pending.

Fixes Manifest_MultiPeriod_1080p.mpd some more.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/2838>
2022-08-04 13:43:57 +01:00