The texture client bound in a GPUVideoImage can be destroyed underneath
it if the GPU process crashed and was restarted. This is because the
ImageBridgeChild singleton will destroy any textures bound to it before
reinitializing itself. If there is a race to access a now stale
GPUVideoImage object, we could accidently dereference the underlying
TextureData which is now null in the TextureClient which was kept alive
in the GPUVideoImage object. Now we check to make sure the TextureData
is still there before accessing.
Automatic update from web-platform-testswpt: Create EventListener-addEventListener test.
This is a follow-up of https://crrev.com/c/1226893 and moves
eventlistener-with-remote-context.html to WPT.
The test content is conceptually the same.
Bug: 883643, 886588
Change-Id: I7c941e162a076d9cbb7abc1d259ff98742a14517
Reviewed-on: https://chromium-review.googlesource.com/1235774
Reviewed-by: Kent Tamura <tkent@chromium.org>
Commit-Queue: Yuki Shiino <yukishiino@chromium.org>
Cr-Commit-Position: refs/heads/master@{#592756}
--
wpt-commits: c8c646139bf7a4dbb78f724f1bd8ce807da3e534
wpt-pr: 13106
Automatic update from web-platform-testsHandle case where mozlog is not installed. (#13113)
In case that mozlog is present we need to ensure any lock it holds is
released in subprocesses. In case it's not present we don't need to do
anything, so don't fail if we can't import it.
Fixes https://github.com/web-platform-tests/wpt/issues/13112.
--
wpt-commits: 1f8fa9e610d934490538812d6121a77baf646bfc
wpt-pr: 13113
Automatic update from web-platform-tests[Background Fetch] Miscellaneous cleanups.
1. A check for state needs to be removed from a WPT test, since we've removed this
attribute from the BackgroundFetchRegistration object.
2. Update calls in matchImpl() from resolver->Reject to ScriptState::RejectWithDOMException.
Change-Id: Ie979ad3675ac7d093ae908604e683802db8115a0
Reviewed-on: https://chromium-review.googlesource.com/1230713
Reviewed-by: Peter Beverloo <peter@chromium.org>
Commit-Queue: Mugdha Lakhani <nator@chromium.org>
Cr-Commit-Position: refs/heads/master@{#592050}
--
wpt-commits: 8abd5d0c3f43ebd83c533e6ac096686879183bd8
wpt-pr: 13051
Automatic update from web-platform-testsFetch: Object.prototype shouldn't interfere with fetch streams
Test that setting Object.prototype.start to a function that throws
doesn't interfere with fetch operations that create a stream. Similarly,
verify that setting throwing accessors for 'type', 'size', and
'highWaterMark' doesn't cause problems.
See https://github.com/whatwg/fetch/pull/781 for background.
--
wpt-commits: ba582388f4506ea2c8341dac4d8ce6e5eb39871b
wpt-pr: 12350
Automatic update from web-platform-tests[css-ruby] Parsing and inheritance tests
Test that ruby-align and ruby-position are inherited and support
the full grammar.
https://drafts.csswg.org/css-ruby-1/#property-index
--
wpt-commits: 0be8bfbbead7340522425c5a9b1dde51588666b0
wpt-pr: 13087
Automatic update from web-platform-testsMerge pull request #11847 from ewilligers/animatedPoints-not-animated
Test that animatedPoints during a SMIL animation do not reflect the animated values (per SVG2 spec)
--
wpt-commits: 8350d8300e729f804135350be205cdcec4af3867
wpt-pr: 11847
Automatic update from web-platform-testsAllow any blockified display type for LEGEND.
Fixed one test that I expected to pass, but didn't because it was
pointing to the wrong ref. Fixed that, and now it passes. Updated
some expectation files. More stuff from various tests (but not
whole tests) is passing than before.
Bug: 781135
Change-Id: Ieb1312b8bf13c7a45afe644ad2a58f3f955044b6
Reviewed-on: https://chromium-review.googlesource.com/1228073
Reviewed-by: Emil A Eklund <eae@chromium.org>
Commit-Queue: Morten Stenshorne <mstensho@chromium.org>
Cr-Commit-Position: refs/heads/master@{#592403}
--
wpt-commits: caca00d31b517ca00d84e19a84a92119836a211a
wpt-pr: 13071
Automatic update from web-platform-tests[css-grid] Property initial values and inheritance (#13045)
Initial values of CSS Box Alignment properties match spec
https://drafts.csswg.org/css-grid/#property-index
None of the properties are inherited.
--
wpt-commits: cc8e7ca0d2f1eb64851cd15c72958c84e05ae93e
wpt-pr: 13045
Automatic update from web-platform-tests[css-text-decor] Inheritance and initial values (#13058)
Test that CSS Text Decoration properties inherit or not
according to spec. Test that they have the expected
initial values.
https://drafts.csswg.org/css-text-decor-3/#property-index
Fix asserts in inheritance-testcommon.js
--
wpt-commits: 8a83c208d9108f3c99dd266a219438bc5adef953
wpt-pr: 13058
Automatic update from web-platform-tests[css-align] Property inheritance, initial values (#13044)
CSS Box Alignment properties are not inherited.
The initial values match the spec
https://drafts.csswg.org/css-align-3/#property-index
--
wpt-commits: fa1cc47c238842580860c04f29a0ddc6d0a575f9
wpt-pr: 13044
Automatic update from web-platform-testsHTML: test different block margins on legend
--
wpt-commits: 50d118302bbe8c44d44fff6d36cc514ae8e311af
wpt-pr: 13054
Automatic update from web-platform-tests[css-layout-api] Add test for percentage height in quirks mode.
Adds a test to verify that constraints.fixedBlockSize is set to the
correct value, in quirks mode, with a percentage block-size.
Bug: 726125
Change-Id: I97a1e946cdbb4443d70803bfd4ace7f8d3df2ce1
Reviewed-on: https://chromium-review.googlesource.com/1231385
Reviewed-by: Aleks Totic <atotic@chromium.org>
Reviewed-by: Morten Stenshorne <mstensho@chromium.org>
Commit-Queue: Morten Stenshorne <mstensho@chromium.org>
Cr-Commit-Position: refs/heads/master@{#592181}
--
wpt-commits: f1feab0fce77d29d1c7eb5a043d2345e6ba3123f
wpt-pr: 13055
- Refactor the highlighter shown/hidden event handlers from MarkupElementContainer in ElementEditor.
We do this because we want to be a bit more lazy when we set these on/off highlighters to only when
there is only a display badge and appropriate display type.
- This fixes a flickering bug cause by the event handlers when you click on the display badges in the markup view.
STR:
1. Have 2 grids shown on the markup view.
2. Toggle on one of them
3. Toggle the other one.
4. See that the active state toggles on, off and on.
We want to toggle the active state manually instead of waiting for the highlighter event handler
to toggle its active state when we click on the display badge. This is because waiting for
the highlight event handler to toggle the active state will cause a brief delay from showing
the active state when the user clicked on the display badge. To fix this, we added a way
to start/stop the highlighter event handlers.
Summary:
This method isn't specific to tabs. It can attach to any "target" actor that inherits from BrowsingContextActor.
Depends On D6161
Reviewers: yulia!
Tags: #secure-revision
Bug #: 1485676
Differential Revision: https://phabricator.services.mozilla.com/D6162
MozReview-Commit-ID: Exz5fWyWyfN
Summary:
Now that all the "remoting" of this method has been moved to TargetFactory.createTargetForTab,
we should rename this method to what it does now. It mostly call attach requests
of the target actor and its child console actor.
It also "connect" the webextension target actor, but I would like to eventually move that
outside of TabTarget.attach, like makeRemote.
Depends On D4078
Reviewers: yulia!
Tags: #secure-revision
Bug #: 1485676
Differential Revision: https://phabricator.services.mozilla.com/D6161
MozReview-Commit-ID: KmFi1LIUBga
Summary:
When switching to async, it is important to catch exception or register a rejection handler
so that errors keep being logged.
So in this patch I'm catching exception in a couple of important codepath.
Depends On D4541
Reviewers: yulia!
Tags: #secure-revision
Bug #: 1485676
Differential Revision: https://phabricator.services.mozilla.com/D4542
MozReview-Commit-ID: IDPJVkAPbTs