Those might be temporary issue, for example s3 returns SERVICE_UNAVAILABLE high
load, but afterward a few tries the request will work, and it has been observed
that internal server error sometimes "fix themselves" so it makes sense to
also retry requests, in case.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8128>
Add administrative REST API endpoints to RangeHTTPServer allowing tests to:
- Configure HTTP return codes for specific paths:
* PUT /admin/status-rules to set rules
* Support time-based expiry with "during" parameter
* Support count-based expiry with "repeat" parameter
* DELETE /admin/status-rules/<path> to remove rules
- Track failure statistics:
* PUT /admin/failure-counts/start to begin monitoring
* GET /admin/failure-counts/<path> to get current count
Useful for testing HTTP retry mechanisms, error handling and failure
recovery behaviors.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8128>
For some cases, maybe the timestamp of gap event is smaller than
segment start value or larger than segment stop value in playback.
And the timestamp plus duration may exceed segment boundary. Need
check and clip the timestamp or duration before recalculating.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8517>
We cannot rely only on the DISCONT flag when deciding whether we need to
request a new key frame or not because it might be that the packet that just
came in with the DISCONT flag is actually the start of a keyframe.
This patch improves the logic to be the same as rtph265depay, by only requesting
a key frame if the packet with the DISCONT flag is not the first one of a FU; or
if its the first one, only request it when we could drop packets due to a
missing key frame.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8343>
I spent too much time trying to navigate validate Python code to figure
out where a test was coming from. Hoping that it's slightly easier for
the next person, this patch:
* Adds type annotations to setup_tests(), for the sake of code
navigation.
* Adds comments matching each test generator with the patterns of test
names it produces.
* Removes an if statement in `register_default_scenarios()` where both
branches have the same exact code with the same exact very long list.
* Removes NamedDic [sic] and replaces it with SimpleNamespace from the
standard library (3.3+) which has the same purpose and API.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8090>
When the driver prefered format is not picked by downstream, the
decoders needs to select another format from the list. The selection
was currently unsorted, resulting in 10bit data often being stripped
to 8bit.
To solve this, reorder the formats in an HW preference order. This order
deviates slightly from the preferred order in libgstvideo. This is to
prefer bandwidth saving over better CPU alignment. As an example NV15 is
prefered over P010. We also prefer tiled over linear.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8522>
Scenario is using wayland with this pipeline: videotestsrc ! glupload ! qml6glsink.
First pipeline and qml construction works just fine. However if all GStreamer
and QML scenegraph resources are removed, GStreamer will call eglTerminate() and
sever the connection to the display server for Qt. When Qt attempts to do any
further GL operations like construct a new QML scene, it can crash in any number
of places as libEGL will start returning NULL or other unexpected values.
What we really need is to ensure that if an external API (e.g. Qt) will
eventually call eglTerminate(), there is no need for GStreamer to call
eglTerminate(). This is what the foreign display flags allows setting.
There is also another possible scenario where one may like to make GStreamer
assume ownership of an EGLDisplay and thus eventually call `eglTerminate()`. As
such, it is now also possible to mark an GstGLDisplayEGL as non-foreign.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8504>
Add new property to support application allocated GstCudaMemory.
CUDA memory alloc/free is a global device synchronization point
as if launching CUDA kernel on default CUDA stream. To avoid the global
synchronization, we added stream-ordered allocation support
which allocates CUDA memory asynchronously.
However, NVENC does not allow registering the stream-ordered
allocated memory. Thus encoder was allocating normal CUDA
memory in case that input CUDA memory is stream-ordered type.
In this commit, newly introduced property will allow application
to provide encoder with GstCudaBufferPool. Application can
preallocate sufficient amount of CUDA memory in advance
to avoid global device synchronization while pipeline is running.
For now, this pool is used only if input CUDA memory is allocated
via stream-ordered-allocation
Part-of: <https://gitlab.freedesktop.org/gstreamer/gstreamer/-/merge_requests/8516>
