Avoid deadlock when closing a pipeline.
<!-- Please describe your changes on the following line: -->
At the moment, the constellation blocks on a pipeline during closure. This PR makes pipeline closure asynchronous.
---
<!-- 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#11546.
- [X] These changes do not require tests because testing for absence of deadlock is difficult.
<!-- 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="35" align="absmiddle" alt="Reviewable"/>](https://reviewable.io/reviews/servo/servo/11585)
<!-- Reviewable:end -->
WebRender.
This happens asynchronously, just as it does in non-WebRender mode.
This functionality is a prerequisite for doing proper display-list-based
hit testing in WebRender, since it moves the scroll offsets into Servo
(and, specifically, into the script thread, enabling iframe event
forwarding) instead of keeping them private to WebRender.
Requires servo/webrender_traits#55 and servo/webrender#277.
Partially addresses #11108.
Always include the last-resort font
This is used as a fallback for any characters that don't have glyphs in the specified font. Without this, per-glyph font fallback doesn't work because the FontGroup always contains only one font. Fixes missing glyphs on many pages on my Linux box.
As a follow-up, we should probably have a smarter strategy for finding fallback fonts, possibly varying by script. (Currently we just have a few hard-coded family names.)
r? @glennw
<!-- Reviewable:start -->
---
This change is [<img src="https://reviewable.io/review_button.svg" height="35" align="absmiddle" alt="Reviewable"/>](https://reviewable.io/reviews/servo/servo/11303)
<!-- Reviewable:end -->
Changes include:
- Introduce an IpcSend trait to abstract over a collection of IpcSenders
- Implement ResourceThreads collection to abstract the resource-related
sub threads across the component
- Rename original ResourceThread and ControlMsg into an unifed CoreResource__
to accommodate above changes and avoid confusions
It's a pointless abstraction that propagates the obsolete chan terminology,
swaps the order in which the sender and receiver are returned, and hides a
source of panics.
