stylo: Improve restyling performance
This commit adds hooks to the Servo style traversal to avoid traversing all the
DOM for every restyle. Additionally it changes the behavior of the dirty flag to
be propagated top down, to prevent extra overhead when an element is dirtied.
This commit doesn't aim to change the behavior on Servo just yet, since Servo does extra job when dirtying the node related with DOM revision counters that might be necessary.
CC @asajeffrey for the DOM revision counters stuff. When a node is dirty, do all its descendants really need to increment the revision counter, or is this an unintended effect? My intuition is that this is hurting performance quite a lot for servo.
r? @bholley
<!-- Please describe your changes on the following line: -->
---
<!-- Thank you for contributing to Servo! Please replace each `[ ]` by `[X]` when the step is complete, and replace `__` with appropriate data: -->
- [x] `./mach build -d` does not report any errors
- [x] `./mach test-tidy` does not report any errors
<!-- Either: -->
- [x] These changes do not require tests because no geckolib tests yet.
<!-- Pull requests that do not address these steps are welcome, but they will require additional verification as part of the review process. -->
<!-- Reviewable:start -->
---
This change is [<img src="https://reviewable.io/review_button.svg" height="34" align="absmiddle" alt="Reviewable"/>](https://reviewable.io/reviews/servo/servo/12563)
<!-- Reviewable:end -->
Fix backward navigation
make use of history iterator
Add frame iterator
add different back logic
cleanup navigation_info
Add extra explanation for iter logic
Remove forward history on full frame tree
Rename navigation to traversal where appropriate
check full tree for can go back/forward
simplify frame iter logic
remove FrameIterator
cleanup history iter
reduce amount of vec allocations
removed extra parenthesis
Remove history iterator
cleanup after rebasing
avoid recursive vec allocation
remove full_frame_tree
remove_forward_history_in_frame_tree -> clear_joint_session_future
Allow window elements as well as iframes to the the target of mozbrowser events
<!-- Please describe your changes on the following line: -->
Allow mozbrowser events, in particular mozbrowsererror events, to target a window. Needed for https://github.com/browserhtml/browserhtml/issues/1182
---
<!-- Thank you for contributing to Servo! Please replace each `[ ]` by `[X]` when the step is complete, and replace `__` with appropriate data: -->
- [X] `./mach build -d` does not report any errors
- [X] `./mach test-tidy` does not report any errors
- [X] These changes fix#12420
- [X] These changes do not require tests because we're not testing our issue reporting system, which this is intended for.
<!-- Pull requests that do not address these steps are welcome, but they will require additional verification as part of the review process. -->
<!-- Reviewable:start -->
---
This change is [<img src="https://reviewable.io/review_button.svg" height="34" align="absmiddle" alt="Reviewable"/>](https://reviewable.io/reviews/servo/servo/12426)
<!-- Reviewable:end -->
We needed a separate thread in the chrome process because communication to the
compositor is done through a trait object, and cross-process virtual calls are
forbidden.
Also, the fact that these messages are ultimately handled by the compositor is
an implementation detail; conceptually, the relevant constellation is supposed
to handle these messages.
So instead, the script thread will now send the messages to the constellation,
which will ask the compositor to handle them.
Fixed conflict
Fixed merge issue
Finished implementation
Disable tinyfiledialogs on Windows
addressed comments
Use ancestor's SubpageId
Move display alert from method to function
Add extra test for nested iframes
Addressed comments
Updated tinyfiledialogs
This allows keeping the VertexAttrib* calls asynchronous.
Another option would be to do the validation in the apply() function,
but that'd require us passing an unnecessary channel around and add
extra synchronization.
The counterpart of this is that it has to be updated when the context
changes, but that's less problem.
This changes headless operation to strictly be a runtime option, rather
than a compile-time one. Note that the old headless version still relied
on a display server to support WebGL, while it now requires one all the
time.
Fixes#8573
Also change when pipelines become active.
This makes the constellation activate a pipeline as the current frame
when it is ready to do initial reflow, rather than when it is ready
to paint.
This fixes a number of intermittent failures that could previously occur
if an iframe was not visible - which would mean it was never moved from
a pending frame in the constellation to an active frame.
(It happens that webrender exposes these intermittents as permanent failures).
