Commit graph

150 commits

Author SHA1 Message Date
Hou Qi
95ac8b0cea decodebin3: filter error message and store latency message for candidate decoder
If the GST_MESSAGE_SRC of error message belongs to candidate decoders,
filter the error message and don't forward it as there might be a
following candidate decoder that can be used.

If the GST_MESSAGE_SRC of error message belongs to candidate decoders,
store the latency message and handle it after decoder is accepted.
This is to avoid the selection lock failure if decodebin3 needs to
handle latency message for candidate decoders when sending sticky event.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/4677>
2023-06-02 14:51:38 +00:00
Hou Qi
887ae4d9e0 decodebin3: try candidate decoders to select the first one that works
Send sticky events to the new created decoder after it switches
to PAUSED state. It it fails, just skip this decoder and try the
next one until finding one that works. Otherwise remove this
failing stream after trying all decoders and no one can work.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/4677>
2023-06-02 14:51:38 +00:00
Hou Qi
6fc6e934aa decodebin3: send sticky event to decoder after setting it to PAUSED
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/4677>
2023-06-02 14:51:38 +00:00
Hou Qi
837169a221 decodebin3: add function remove_decoder_link()
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/4677>
2023-06-02 14:51:38 +00:00
Hou Qi
536c344111 decodebin3: copy sticky event
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/4677>
2023-06-02 14:51:38 +00:00
Edward Hervey
9befb81036 urisourcebin: Set source element to READY before querying it
Generating the source element is done when urisourcebin is doing the READY to
PAUSED state change, so it is reasonable to set the new source element to that
state.

This also allows detecting early failures with backing libraries or
hardware (checks done in NULL->READY).

Finally it makes more sense to have an element in READY when attempting to query
information from it (such as SCHEDULING queries or probing live-ness).

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3856>
2023-06-02 07:00:43 +00:00
Guillaume Desmottes
b3d27da397 streamsynchronizer: check reset-time when handling FLUSH_STOP
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/4749>
2023-06-01 09:45:50 +02:00
Guillaume Desmottes
c2d8f4f729 streamsynchronizer: reset eos on STREAM_START
self->eos was never reset after streamsynchronizer has sent EOS
(except on explicit flush or switching back to PAUSED).
As a result, synchronization was broken if new streams were pushed later
as gst_stream_synchronizer_wait() does not wait if self->eos is set.

Fix this by reseting self->eos on STREAM_START as that means a new
stream is being sent upstream and so a new EOS will follow later on.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/4749>
2023-06-01 09:45:15 +02:00
Edward Hervey
d6f1c517f3 decodebin3: Handle changes in stream type
While decodebin3 could handle changes in inputs (ex: changing codecs), there was
still one limitation which was when changing between sources which had
non-intersecting stream types (ex: switching from a video-only source to a
audio-only source). While the decoder *could* change to the proper codec ... it
would carry on using a `DecodebinOutputStream` associated to that stream
type (and therefore with pads with the wrong name).

In order to handle this:

* We notify the `MultiQueueSlot` of the change in `GstStreamType` if it already
  had an associated inputstream (ex: the one associated with the static sink
  pad)

* We detect such changes on the output of multiqueue as soon as
  possible (i.e. when we get the GST_EVENT_STREAM_START for the new stream type)
  by discarding the associated output.

Fixes https://gitlab.freedesktop.org/gstreamer/gstreamer/-/issues/1669

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/4703>
2023-05-25 21:23:21 +00:00
Edward Hervey
f51283b57b uridecodebin3: Also re-use decodebin3 static sink pad
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/4703>
2023-05-25 21:23:21 +00:00
Guillaume Desmottes
58f80c180c uridecodebin3: notify when 'current-(sub)uri' properties are updated
Assume both uri and suburi are changed when the main item changes.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/4543>
2023-05-24 09:28:58 +00:00
Haihua Hu
9f67b866b9 decodebin3: avoid identity, sinkpad, parsebin leakage when reset input
when reset_input, need remove identity/parsebin from decodebin3
when release_pad, need call free or reset input if collection
didn't change

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/4664>
2023-05-19 16:24:08 +00:00
Haihua Hu
36333a5152 decodebin3: fix random hang when remove failing stream
When reconfigure_output_stream entry missing decoder path,
requested_selection should been update with what is really
active/selected immdiately with SELECTION_LOCK hold. So
use an optional message return from reconfigure_output_stream
and post it after release SELECTION_LOCK. This can make sure
other thread call to check_slot_reconfiguration will got
a correct requested_selection.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/4599>
2023-05-16 12:43:12 +00:00
Edward Hervey
0db6149880 decodebin3: Handle streams without CAPS or TIME segment
decodebin3 will do its best to figure out whether a parsebin is required to
process the incoming stream.

The problem is that for push-based stream it could happen that the stream would
not provide any caps, resulting in nothing being linked internally.

Furthermore, there is the possibility that a stream *with* caps would not be
using a TIME segment, which is required for multiqueue to properly work.

In order to fix those two issues, we force the usage of parsebin on push-based
streams:
* When the pad is linked, if upstream can't provide any caps
* When we get a non-TIME segment

Fixes #2521

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/4492>
2023-04-27 12:50:21 +00:00
Hou Qi
aed4d31e67 decodebin3: fix segment fault when print decoder log
Segment fault happens when cannot find decoder but try to print
decoder name. Need to check the decoder.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/4461>
2023-04-25 21:42:15 +08:00
Patricia Muscalu
c6bb20bba8 playsink: Fix volume leak
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/4459>
2023-04-20 14:21:15 +00:00
Haihua Hu
8541d6b990 decodebin3: fix hang issue when remove failing stream
Need mark selection_update to true when update selection,
otherwise, pipeline will not handle this selection update
sometimes when this flag has been reset

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/4277>
2023-03-27 13:52:42 +00:00
Edward Hervey
0be233a446 uridecodebin3: Remove play items that were never connected
This is a follow-up of the previous commit that enabled support for redirection.

The problem is that the urisourcebin that emitted the error redirection never
produced any pads, and therefore was never linked to decodebin3. This resulted
in the code waiting for that (output) item to finally switch over ... which will
never happen.

The fix is done by removing it early if it was never connected to decodebin3.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/4252>
2023-03-23 07:39:48 +01:00
Edward Hervey
133c4b2ff9 uridecodebin3: Handle redirection errors
This is done by doing an immediate switch to the redirection URI if compatible.

Fixes #1562

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/4242>
2023-03-22 11:28:47 +00:00
Edward Hervey
095356b67e parsebin: Improve elementary stream handling
The goal of parsebin is to figure out which elements to link together in order
to provide elementary streams given any random input.

The problem is that deciding whether a given stream should still have more
elements plugged in or not was dependent on ... the presence of compatible
decoders (sic).

Instead of that, if we can't plug anymore elements on a given stream *and* it is
detected as being an elementary stream, expose it.

Fixes #2118

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/4231>
2023-03-21 07:26:37 +00:00
Edward Hervey
cf3adbcfb3 urisourcebin: Activate pad before transferring sticky events
Otherwise they get refused since the pad is flushing

Fixes #2384 for good

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/4226>
2023-03-20 18:11:26 +00:00
Edward Hervey
7f1c5e2ef3 urisourcebin: Propagate sticky events from parsebin
If sticky events are present on parsebin source pads, we propagate them to the
multiqueue source pads. Those will be propagated on the new urisourcebin source
pads like in the other code paths.

This ensures that STREAM_START event are present on new source pads. If CAPS
event are also present (not guaranteed), they will also be available.

Fixes #2384

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/4203>
2023-03-17 23:48:43 +00:00
Tim-Philipp Müller
8759b77a50 gst-plugins-base: re-indent with GNU indent 2.2.12
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/4182>
2023-03-17 03:18:53 +00:00
Edward Hervey
b4cadf35bd uridecodebin3: Ensure atomic urisourcebin state change
When dynamically adding and synchronizing the state of urisourcebin, we need to
ensure that no-one else attempts to change the state in case of failures

Fixes https://gitlab.freedesktop.org/gstreamer/gstreamer/-/issues/1803

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/4125>
2023-03-08 09:29:57 +00:00
Thibault Saunier
4067bbdd91 uridecodebin: Set source element to READY before querying it
Generating the source element is done when uridecodebin is doing the
READY to PAUSED state change, so it is reasonable to set the new source
element to that state.

This also allows detecting early failures with backing libraries or
hardware (checks done in NULL->READY).

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3857>
2023-02-08 11:07:21 +00:00
Tim-Philipp Müller
165fdac5c6 playback: drop use of GSlice
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3695>
2023-01-24 15:25:06 +00:00
Edward Hervey
4c1d01ff57 playback: New playback elements are no longer experimental
The API is not expected to change and it has been used extensively since 1.18.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3761>
2023-01-21 12:28:21 +00:00
Edward Hervey
e2773304a2 decodebin3: Remove failing stream from selection
If no decoder is present for a given stream, remove it from the requested
selection.

Fixes #1581 and #1662

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3754>
2023-01-20 08:54:00 +00:00
Edward Hervey
34ea792881 decodebin3: Don't intercept queries if no parsebin present
If we don't use a parsebin, we forward the queries as-is from upstream. There is
no reconfiguration possible within identity.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3754>
2023-01-20 08:54:00 +00:00
Edward Hervey
1a30ed8113 decodebin3: Don't insert parsebin if input is already parsed
This is a temporary workaround until we find a generic solution to indicate that
a stream has already been "parsed".

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3754>
2023-01-20 08:54:00 +00:00
Tim-Philipp Müller
64c4bfdf7e uridecodebin3: clear subitem if suburi is set to NULL
Fixes https://gitlab.freedesktop.org/gstreamer/gstreamer/-/issues/1649

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3730>
2023-01-18 11:43:31 +00:00
Edward Hervey
13ad8c8e4e urisourcebin: Cleanup on failure
This ensures there is a source present/valid *ONLY* in PAUSED and above.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3720>
2023-01-13 16:41:53 +01:00
Edward Hervey
1546ab0d0f urisourcebin: Unset locked state
Some elements still had it in failure cases

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3720>
2023-01-13 16:41:53 +01:00
Edward Hervey
f444e62672 uridecodebin3: Handle state synchronization failures
Properly reset and remove the failing urisourcebin if we failed bringing them up
to the same state as ourselves.

Fixes https://gitlab.freedesktop.org/gstreamer/gstreamer/-/issues/1708

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3720>
2023-01-13 16:41:53 +01:00
Edward Hervey
1fe99ebae1 decodebin3: Fix global group-id handling
The goal of the "global" group-id is to fix new inputs that do not come from the
same "source" as others. In order to ensure all "current" streams have the same
group-id we distribute the first valid group-id to all streams.

This commit fixes two issues with that:

* When inputs are unlinked they weren't always properly resetted (it would only
  work if parsebin is used, which is no longer the default in
  uridecodebin3/playbin3).

* When computing the global group-id, take into account unset
  group-id (i.e. GST_GROUP_ID_INVALID).

Fixes https://gitlab.freedesktop.org/gstreamer/gstreamer/-/issues/1698

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3712>
2023-01-11 23:29:05 +00:00
Guillaume Desmottes
22917b140f decodebin3: ensure recalculate_group_id() is called with INPUT lock
This function uses main_input and current_group_id which are protected
by the input lock.
Some calls were already safe but not all.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3702>
2023-01-11 18:37:05 +00:00
Guillaume Desmottes
cad0768191 uridecodebin3: fix stream leak
GstSourcePad owns the GstStream but was not unreffing it when being
disposed.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3617>
2023-01-11 16:44:21 +00:00
Guillaume Desmottes
da696477c6 decodebin3: fix dead lock when removing pad
gst_element_remove_pad() is triggering a call to
gst_decodebin3_input_pad_unlink() which needs the input lock as well,
resulting in a dead lock.

Fix #1667

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3614>
2023-01-11 15:06:37 +00:00
Edward Hervey
6beca9f055 urisourcebin: Fix parsebin handling
The goal of !3601 was to make sure we don't end up with non-streams-aware
adaptive demuxers. Since we know that parsebin is streams-aware, just remember
that and handle them that way.

Fixes issues with some scenarios where parsebin might add pads *before* the
collection is posted on the bus

Fixes https://gitlab.freedesktop.org/gstreamer/gstreamer/-/issues/1675

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3658>
2022-12-30 08:46:14 +01:00
Edward Hervey
66b96baf23 uridecodebin3: Fix handling of "expected" pads
The number of expected pads was:
* Defaulting to 1
* Or being overriden by GST_MESSAGE_STREAMS_SELECTED

This fails if upstream isn't a selectable source and has multiple streams, and
would therefore cause failures with multi-stream gapless playback

Fixes #1672

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3658>
2022-12-30 08:46:14 +01:00
Edward Hervey
80e7c31514 uridecodebin3: Avoid repeatedly calling the blocking probe
When skipping an event, we want to unref it and say we handled it. This avoids
being repeatedly called for the same (sticky) events.

The events will be properly propagated once the pad is linked.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3658>
2022-12-30 08:46:13 +01:00
Edward Hervey
a0655d5131 uridecodebin3: Handle multiple pending pad block probes
It is quite possible to have the blocking probe called from different streaming
threads when all expected pads are present.

* Notify all waiters by using g_cond_broadcast instead of g_cond_signal
* Properly remove the probe after waiting

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3658>
2022-12-30 08:46:13 +01:00
Edward Hervey
74a503c197 urisourcebin: Modify check for streams-aware adaptive demuxers
Using the "GstBin" flags to check if an adaptive demuxer is streams-aware isn't
a good idea since it prevents using elements which aren't bins.

Instead we see if a collection was posted by the demuxer by the time a pad is
added.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3601>
2022-12-19 15:23:32 +00:00
Guillaume Desmottes
1f929ba32f uridecodebin3: fix typo in doc
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3602>
2022-12-19 14:24:06 +01:00
Edward Hervey
36c3bb0d05 uridecodebin3: Protect against NULL uri/suburi
Fixes #1625

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3511>
2022-12-02 16:55:05 +00:00
Edward Hervey
41017e6c34 uridecodebin3: Always store pending buffering messages
Fixes #1586

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3487>
2022-12-02 09:22:41 +00:00
Edward Hervey
3a63eab2fa uridecodebin3: propagate about-to-finish for new inputs
If the pending next urisourcebin is already draining, make sure we propagate it
when activating it.

Fixes gapless playback of small entries

Fixes https://gitlab.freedesktop.org/gstreamer/gstreamer/-/issues/1588

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3457>
2022-11-23 12:19:22 +00:00
Edward Hervey
def2515ae2 uridecodebin3: Remove old fixme
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3457>
2022-11-23 12:19:22 +00:00
Edward Hervey
734a4554f3 (uri)decodebin3: Ensure group-id consistency
Make sure that group-id of a given play item are made consistent from the
start (sources) and all the way through the output.

This ensures that we can reliably detect that we have switched to the next play
item on the output of decodebin3 (and we can therefore properly free/release it)

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3457>
2022-11-23 12:19:21 +00:00
Edward Hervey
353691602e uridecodebin3: Fix shutdown procedures in probe
When shutting down, we want to remove the urisourcebin blocking probes ... but
we also want to propagate a GST_FLOW_FLUSHING upstream (and not
GST_FLOW_NOT_LINKED) to make the upstream task gracefully stop instead of
posting an error message.

Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/3457>
2022-11-23 12:19:21 +00:00