The regressing bug added an assertion that the nursery is enabled when minor GC
is requested. That holds in normal operation but this testing function triggered
it.
Differential Revision: https://phabricator.services.mozilla.com/D180846
From the server, when traversing parent rules, we now do include non at-rules
that have a `selectorText` property, in order to handle CSS nesting.
On the client, we display those selectors as we do for ancestor at-rules (at least
for now, we might add brackets and proper indentation in a follow up).
A test with various nesting is added to check we do display the ancestor rules
as expected.
Differential Revision: https://phabricator.services.mozilla.com/D181041
Currently, this function warns any time it fails to return something.
However, this warning happens extremely frequently in two cases:
when the property is unknown or when somebody tries to get a
profile directory in a child process.
This patch eliminates those. It also consolidates some profile
cases. Hopefully an additional 2 strcmps on common cases won't
matter.
Differential Revision: https://phabricator.services.mozilla.com/D181007
The `extensions.databaseSchema` pref is used by the implementation to
track the migration state of the extension database. This is an internal
implementation detail that doesn't affect external behavior, and cannot
be controlled by individual unit tests.
TV / `--compare-preferences` detects changes in some tests on Android,
without it being actionable. This adds the pref to ignorePrefs.json to
avoid noise.
Differential Revision: https://phabricator.services.mozilla.com/D181095
* This patch expands this keyboard shortcut to support whichever of the three
last actions were taken - last tab closed, last window closed or session restore.
* Existing test file was renamed and additional cases added.
Differential Revision: https://phabricator.services.mozilla.com/D178965
We shouldn't block the iframe if the initial request doesn't have valid
headers, we should only check the final request.
Differential Revision: https://phabricator.services.mozilla.com/D180485
In GfxInfo, probe all V4L2 devices to see which of them can do M2M
hardware decode acceleration. Set the hardware decode support features
appropriately depending on what we find.
Differential Revision: https://phabricator.services.mozilla.com/D178780
Extend the FFmpeg video decoder's hardware acceleration support to
include V4L2 in addition to VAAPI. If hardware acceleration is enabled
and VAAPI is not available then attempt to initialise a V4L2-wrapper
codec (by name, because libavcodec does not correctly prioritise these
codecs as hardware-accelerated). If V4L2 is not available then this
step will fail and we fall back to software decoding.
DRM-PRIME buffer format is used for zero-copy decoding in a similar
manner to the VAAPI decoder, except that FFmpeg implements V4L2 using
hardware wrapper decoders instead of using a hwaccel so we don't store a
hwcontext.
This adds a use of the libavcodec avcodec_find_decoder_by_name function,
so I add this to the ffmpeg library wrapper.
Differential Revision: https://phabricator.services.mozilla.com/D178763
Adding V4L2 hardware acceleration support for video playback requires
the RDD to be allowed to:
- Look in /dev to enumerate available V4L2 devices
- Open /dev/video* (but only m2m devices, not capture ones)
- Attempt to make sockets to connect to the wayland server (we can deny
the requests to make a socket, just don't kill the process for trying).
- Use VIDEO-type ioctls
Differential Revision: https://phabricator.services.mozilla.com/D178194
Adds a PipeWire based camera support that was recently merged into
WebRTC. This should be an experimental feature for now and therefore it
is kept behind a config option.
Differential Revision: https://phabricator.services.mozilla.com/D176625
Wpt test failures for Android relase in the following tests seem being fixed by 1833652.
Update the expectation files accordingly.
testing/web-platform/html/semantics/popovers/popover-inside-display-none.html
testing/web-platform/html/semantics/popovers/popover-open-display.html
testing/web-platform/html/semantics/popovers/popover-open-overflow-display.tentative.html
Differential Revision: https://phabricator.services.mozilla.com/D181054
The modified test, helper_long_tap.html, is mostly now same as the Android one.
A difference is on the Linux/Mac version we observe a mouselongtap event after a
contextmenu event wasn't preventDefaulted.
Depends on D180636
Differential Revision: https://phabricator.services.mozilla.com/D180637
`mouselongtap` events are internally used for accessible caret on desktops.
Whether to fire a touchcancel event here should just depend on whether a context
menu opens or not.
Differential Revision: https://phabricator.services.mozilla.com/D180636
Adds a PipeWire based camera support that was recently merged into
WebRTC. This should be an experimental feature for now and therefore it
is kept behind a config option.
Differential Revision: https://phabricator.services.mozilla.com/D176625
