Use the object lock for the following fields:
- `bytes`: Written by the `load-bytes` signal unless running; consumed
on start.
- `draw_background`: Read and written by the `draw-background`
property.
- `location`: Read and written by the `location` property and the URI
handler.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/2222>
Problem is that unreffing the EGLImage/SHM Buffer while holding the
images_mutex lock may deadlock when a new buffer is advertised and
an attempt is made to lock the images_mutex there.
The advertisement of the new image/buffer is performed in the
WPEContextThread and the blocking dispatch when unreffing wants to run
something on the WPEContextThread however images_mutex has already been
locked by the destructor.
Delay unreffing images/buffers outside of images_mutex and instead just
clear the relevant fields within the lock.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1843>
Using the object lock is problematic for anything that can dispatch to
another thread which is what createWPEView() does inside
gst_wpe_src_start(). Using the object lock there can cause a deadlock.
One example of such a deadlock is when createWPEView is called, but
another (or the same) wpesrc is on the WPEContextThread and e.g. posts a
bus message. This message propagations takes and releases the object
lock of numerous elements in quick succession for determining various
information about the elements in the bin. If the object lock is
already held, then the message propagation will block and stall bin
processing (state changes, other messages) and wpe servicing any events.
Fixes https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/issues/1490
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1934>
As waiting for the load to be finished is specific to the WebView, it should be
done from our WPEView, not from the WPEContextThread. This fixes issues where
multiple wpesrc elements are created in sequence. Without this patch the first
view might receive erroneous buffer notifications.
Fixes#1386
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1568>
The WPEThreaded view is now split in 2 classes:
- WPEContextThread handles the persistent WebKit thread, where all WebKit API
calls should be handled.
- WPEView: is created from the WPEContextThread. It handles the WebView and
maintains the public interface on which wpesrc relies. This is the facade for
the WebView, basically. It takes care of dispatching API calls into the context
thread.
With these fixes it is now possible to create (and reuse) mutlple wpesrc
elements during the application lifetime.
Fixes#1372
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1484>
Receiving the WEBKIT_LOAD_COMMITTED event doesn't actually
mean we have committed an SHM buffer / image yet.
As this is the condition we are interested in, check it
instead.
Also wrap g_cond_wait in a loop for extra correctness points.
Part-of: <https://gitlab.freedesktop.org/gstreamer/gst-plugins-bad/-/merge_requests/1476>
The previous version of the SHM export support still required a valid
EGLDisplay. The upcoming WPEBackend-FDO 1.8.x aims to remove this requirement,
hence allowing wpesrc to be used without GPU.
Starting from WPEBackend-FDO 1.6.x, software rendering support is available.
This features allows wpesrc to be used on machines without GPU, and/or for
testing purpose. To enable it, set the `LIBGL_ALWAYS_SOFTWARE=true` environment
variable and make sure `video/x-raw, format=BGRA` caps are negotiated by the
wpesrc element.
By passing NULL to `g_signal_new` instead of a marshaller, GLib will
actually internally optimize the signal (if the marshaller is available
in GLib itself) by also setting the valist marshaller. This makes the
signal emission a bit more performant than the regular marshalling,
which still needs to box into `GValue` and call libffi in case of a
generic marshaller.
Note that for custom marshallers, one would use
`g_signal_set_va_marshaller()` with the valist marshaller instead.
If the mutex is locked while running frameComplete there is a potential deadlock
bound to happen when we get a new exported images from the backend.
Fixes#1101
