Since commit 4d86f994, when setting an RTSP media both shared and
reusable, streaming cannot be restarted after the first time all the
clients disconnect. That happens because the sockets (unlike
addresses) of GstRTSPStream are not cleared in
gst_rtsp_stream_leave_bin, and on restart sockets and addresses are
not allocated in gst_rtsp_stream_allocate_udp_sockets, and then the
check in create_sender_part fails. Fix this by clearing sockets in
gst_rtsp_stream_leave_bin.
Fixesgstreamer/gst-rtsp-server#113
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6325>
find_slot_for_stream_id() will return a slot which has the request stream-id as
active_stream *or* pending_stream (i.e. the slot on which that stream is
currently being outputted or will be outputted).
When figuring out which slot to use (if any) we want to consider stream-id
which *will* appear on a given slot which isn't outputting anything yet the same
way as if we didn't find a slot yet.
Fixes races when doing intensive state changes
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6270>
With the new copy_opaque system, the corresponding frame is stored in the
picture opaque ref.
This also handles the case where the "regular" opaque might be empty in the
case of "DECODE_ONLY" frames, since it that field is set in `get_buffer2()`
which might not be called for those frames
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6301>
When we're doing a state change from PLAYING to NULL, first we invoke
gst_rtspsrc_loop_send_cmd_and_wait (..., CMD_CLOSE, ...) during
PAUSED_TO_READY which will schedule a TEARDOWN to happen async on the
task thread.
The task thread will call gst_rtspsrc_close(), which will send the
TEARDOWN and once it's complete, it will call gst_rtspsrc_cleanup()
without taking any locks, which frees src->streams.
At the same time however, the state change in the app thread will
progress further and in READY_TO_NULL it will call gst_rtspsrc_stop()
which calls gst_rtspsrc_close() a second time, which accesses
src->streams (without a lock again), which leads to simultaneous
access of src->streams, and a segfault.
So the state change and the cleanup are racing, but they almost always
complete sequentially. Either the cleanup sets src->streams to NULL or
_stop() completes first. Very rarely, _stop() can start while
src->streams is being freed in a for loop. That causes the segfault.
This is unlocked access is unfixable with more locking, it just leads
to deadlocks. This pattern has been observed in rtspsrc a lot: state
changes and cleanup in the element are unfixably racy, and that
foundational issue is being addressed separately via a rewrite.
The bandage fix here is to prevent gst_rtspsrc_stop() from accessing
src->streams after it has already been freed by setting src->state to
INVALID.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6302>
The transport stream only returned the CAPS for the first matching PT entry
from the `ptmap`. Other SSRC with the same PT where not included. For a stream
which bundled multiple audio streams for instance, only the first SSRC was
knowed to the SSRC demux and downstream elements.
This commit adds all the `ssrc-` attributes from the matching PT entries.
The RTP jitter buffer can now find the CNAME corresponding its SSRC even if it
was not the first to be registered for a particular PT.
The RTP PT demux removes `ssrc-*` attributes cooresponding to other SSRCs
before pushing SSRC specific CAPS to downstream elements.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6119>
And also re-timestamp them with the current buffer's PTS.
Not doing so keeps the timestamps of event packets as
GST_CLOCK_TIME_NONE or the timestamp of the previous buffer, both of
which are bogus.
Making sure that (especially) the first packet has a valid timestamp
allows putting e.g. the NTP timestamp RTP header extension on it.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/5173>
Parse the speed and scale in the server's response
*before* the range, so that the range start/stop
are swapped (or not swapped) correctly based
on the server's actual chosen values. Otherwise,
the old rate from the segment is used - what the
last seek asked for, but not necessarily what
the server chooses.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6248>
In the situation where playback starts from a keyframe before
the target playback segment, then the first buffers will be
outside the configured segment and gst_segment_to_stream_time()
will return GST_CLOCK_TIME_NONE unconditionally.
If drop-out-of-segment is false, the RTP buffers will not be
dropped, but will be sent witout ONVIF extension timestamps
and given GST_CLOCK_TIME_NONE timestamps on the receiver.
Instead, use gst_segment_to_stream_time_full() to extrapolate
stream time outside the segment so that such buffers still
get assigned their correct timestamps on the receiver.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6248>
Don't accidentally include the stuffing byte (if present)
into the bottom field size. It should only be included in the
total segment length.
Fixes problems with FFmpeg not rendering the subtitles
with a stuffing byte, giving a "Invalid object location!" error.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6250>
Cocoa version of glwindow only checks the preferred size upon window creation. glimagesink sets the size right before
calling gst_gl_window_show(), which might be way after the window is created in some cases. If the size was set too
late, glimagesink on macOS would remain 320x240 unless manually resized.
This change makes sure to resize the existing window when _show() is called.
Curiously, this has always been an issue, but went from manifesting every once in a while to being almost completely
broken once old event loop workarounds were removed and gst_macos_main() was introduced.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6185>
Don't use g_return_val_if_fail() to catch the
open-ended segment or empty segment cases in
gst_segment_to_running_time_full()
g_return_val_if_fail() is for programmer errors,
and can be compiled out with a flag.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6219>
Provide a clock from the source that is a monotonic system clock with
the rate corrected based on the measured and ideal capture rate of the
frames.
If this clock is selected as pipeline clock, then provide perfect
timestamps to downstream.
Otherwise, if the pipeline clock is the monotonic system clock, use the
internal clock for converting back to the monotonic system clock.
Otherwise, use the monotonic system clock time calculated in the above
case and convert that to the pipeline clock.
In all cases this will give a smoother time than the previous code,
which simply took the difference between the driver provided capture
time and the current real-time clock time, and applied that to the
current pipeline clock time.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6208>
Otherwise there's a small window between querying the state and doing
the transfer in which a frame could be dropped, and we would then output
the frame right after the dropped one as if it was the dropped frame.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6208>
In low delay B mode, the P frame is converted as B frame with forward
references. For example, One P frame may refers to P-1, P-2 and P-3 in
list0 and refers to P-3, P-2 and P-1 in list1.
So the num in list0 and list1 does not reflect the forward_num and
backward_num. The vaapi does not provide ref num for forward or backward
so far. In this case, we just consider the backward_num to be 1 conservatively.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6249>
In b_pyramid mode, B frames can be ref and prevPicOrderCntLsb can
be the B frame POC which is smaller than the P frame. This can cause
POC diff bigger than MaxPicOrderCntLsb/2 and generate wrong POC value.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6249>
Gets being released memory back to queue even if allocator is flushing
in order to count the number of outstanding memory objects.
Also, clear queue if there's no outstanding memory object and
allocator is flushing
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6240>
Making it possible to properly handle compositors that have those
properties as doubles and handle antialiasing.
Internally we were handling those values as doubles in framepositioner,
so expose new properties so user can set values as doubles also.
This changes the GESFramePositionMeta API but we are still on time for 1.24
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6241>
The `G_DECLARE_FINAL_TYPE` macro does not need to be terminated with a
semicolon and the extra semicolon breaks building e.g. libcamera with
clang because `-Wextra-semi` is used which produces the following
error in conjunction with `-Werror`:
```
gstreamer-1.0/gst/allocators/gstdrmdumb.h:61:43: error: extra ';' outside
of a function is incompatible with C++98 [-Werror,-Wc++98-compat-extra-semi]
61 | GST, DRM_DUMB_ALLOCATOR, GstAllocator);
| ^
1 error generated.
```
Fix this by removing the extra semicolon
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6239>
Syncrhonizing buffer commits to the streaming thread can lead to
dropped frames when frame callbacks are not processed before the
next frame is ready for rendering. Depending on the drift between
the wayland compositor and buffer source timings, this can lead to
periods of significant frame drop, especially when the media frame
rate is close to the display frame rate.
Cache buffers in the streaming thread and peform commits on the
display thread to eliminate the buffer commit racing.
The implementation is the same for both waylandsink and gtkwaylandsink,
so move it to the common wayland library under gst-lib.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6133>
Add synchonized versions of wl_display_sync() and wl_callback_destroy()
that will ensure that to callbacks can be managed in a thread safe way
on the display queue even when they are dispatched from a separate
thread.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6133>
Unprepare method posts WM_GST_D3D11_DESTROY_INTERNAL_WINDOW
command to the window queue, and from that moment considers
internal_hwnd to be released, and so it sets it to null.
The problem is that it's possible that right at that moment
the window thread might be already processing some other
command, or just another command might be already in the queue.
On practice we met a crash when WM_PAINT got processed in between
(unprepare already finished and WM_GST_D3D11_DESTROY_INTERNAL_WINDOW
was not handled yet)
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6187>
In the case where the queue shrinks due to a property change and the queue
becomes full, we would set the waiting_del flag, which would prevent posting the
100% buffering message on the bus. Since the pipeline is not aware of the new
buffering value, in the common case where the pipeline is paused during
buffering, it would never resume.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6165>
Remove the percent_changed check to determine whether a buffering message should
be posted. The check on the last posted buffering value is sufficient, and the
removal doesn't introduce additional complexity.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6165>
If input height and parsed one are identical, do not consider it as interlaced
Fixing below pipeline:
gst-launch-1.0 videotestsrc ! video/x-raw,format=I420,width=640,height=10 \
! jpegenc ! jpegparse ! jpegdec ! videoconvert ! autovideosink
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6181>
When dealing with demuxers which aren't streams-aware, we need to handle the
old-school "stream replacement" dance from `parsebin` and hide that in such a
way that output pads are re-used (if compatible).
By analyzing the collection posted by parsebin, we can:
* Identify whether some output slots are no longer used (because the stream they
currently handle is not present in the collection)
* Decide if some upcoming streams could re-use the existing slot
This supports both buffering and non-buffering modes.
Fixes#1651
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6201>
When the conversion is only caps feature from memory:VAMemory to system memory,
it's possible to optimize by doing a pseudo pass-through since the va-backed
buffers are the same for system memory buffers.
This change will also mitigates #2940
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6174>
If the allocation query received from downstream doesn't handle GstVideoMeta but
it requests memory:DMABuf caps feature, it's incomplete, so we rather reject the
negotiation.
Both in base decoder, base transform and compositor.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6155>
When switching urisourcebin, ensure that we first unlink *all* pads from
decodebin3 before linking them again.
This is to ensure that decodebin3 completely knows that all previous pads are no
longer needed and can prepare itself to being re-used.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6179>
The value is stored as an 8 bit integer, with 0 meaning that there is
not data for this extension. That means that the maximum length is 255
bytes and not 256 bytes.
On the other hand, the one-byte RTP header extensions are storing the
length as a 4 bit integer with an offset of 1 (i.e. 0 means 1 byte
extension length), so here 16 is the correct maximum length.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6180>
This is a simplification of the venerable
gst_va_base_dec_get_preferred_format_and_caps_features() function, which
predates since gstreamer-vaapi. It's used to select the format and the
capsfeature to use when setting the output state. It was complex and hard to
follow. This refactor simplifies a lot the algorithm.
The first thing to remove _downstream_has_video_meta() since, most of the time
it will be called before the caps negotiation, and allocation queries make sense
only after caps negotiation. It might work during renegotiation but, in that
case, caps feature change is uncommon. Better a simple and common approach.
Also, for performance, instead of dealing with caps features as strings, GQuarks
are used.
The refactor works like this:
1. If peer pad returns any caps, the returned caps feature is system memory and
looks for a proper format in the allowed caps.
2. The allowed caps are traversed at most 3 times: one per each valid caps
feature. First VAMemory, later DMABuf, and last system memory. The first to
match in allowed caps is picked, and the first format matching with the
chroma is picked too.
Notice that, right now, using playbin videoconvert never return any.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/6154>
