To actually make the multiprocess communication work, we'll need to
reroute the task creation to the pipeline or the compositor. But this
works as a first step.
profile: Make the time and memory profilers run over IPC.
Uses a couple of extra threads to work around the lack of cross-process
boxed trait objects.
r? @nnethercote
<!-- Reviewable:start -->
[<img src="https://reviewable.io/review_button.png" height=40 alt="Review on Reviewable"/>](https://reviewable.io/reviews/servo/servo/6629)
<!-- Reviewable:end -->
Have BufferMap store NativeSurfaces and rename to SurfaceMap
We currently store LayerBuffers, because previously NativeSurfaces did
not record their own size. Now we can store NativeSurfaces directly,
which saves a bit of space in the surface cache and allows us to create
LayerBuffers only in the PaintTask.
This also means that instead of sending cached LayerBuffers, the
compositor can just send cached NativeSurfaces to the PaintTask.
<!-- Reviewable:start -->
[<img src="https://reviewable.io/review_button.png" height=40 alt="Review on Reviewable"/>](https://reviewable.io/reviews/servo/servo/6679)
<!-- Reviewable:end -->
The idea here is to land this before making images and canvas IPC-safe,
because this will shake out bugs relating to the shared memory. There
are currently test timeouts that are preventing multiprocess images and
canvas from landing, and I believe those are due to the inefficiency of
sending large amounts of data in the unoptimized builds we test with. By
moving to shared memory, this should drastically reduce the number of
copies and `serde` serialization.
Under the hood, this uses Mach OOL messages on Mac and temporary
memory-mapped files on Linux.
We currently store LayerBuffers, because previously NativeSurfaces did
not record their own size. Now we can store NativeSurfaces directly,
which saves a bit of space in the surface cache and allows us to create
LayerBuffers only in the PaintTask.
This also means that instead of sending cached LayerBuffers, the
compositor can just send cached NativeSurfaces to the PaintTask.
Passes some but not all WPT tests. One failure is caused by an issue in codegen for the `DOMPointInit` dictionary, the others by outdated tests: Gecko implements an old version of the spec that overloaded the `DOMPoint` constructor to optionally take an object as the first argument, and made `DOMPointReadOnly` non-constructible.
script: Increase our file descriptor limit on Linux.
We've had problems with this before, and I think it's starting to cause
problems again.
See PRs #6629 and #6616; my current theory is that this is the problem.
r? @larsbergstrom
<!-- Reviewable:start -->
[<img src="https://reviewable.io/review_button.png" height=40 alt="Review on Reviewable"/>](https://reviewable.io/reviews/servo/servo/6707)
<!-- Reviewable:end -->
Optimize `Node.normalize()`.
Do not copy the discarded node's text data, borrow it.
Closes#6658.
p.s. What's the `let text_node = text_node.clone();` for? I removed it because it doesn't seem to be necessary, but I'd like to be sure.
<!-- Reviewable:start -->
[<img src="https://reviewable.io/review_button.png" height=40 alt="Review on Reviewable"/>](https://reviewable.io/reviews/servo/servo/6667)
<!-- Reviewable:end -->
This needs to happen off a task because we won't be able to access the
WebSocket object directly once this code moves to a background thread.
There is no behaviour change, because we make sure that self.ready_state is
not Connecting in Send().
Fix a few issues
- Use SmallVec<[T; N]>
- Make find_iframe a free function
- Make ProgressEvent use enums for bubbles and cancelable
- Change README, as `rust-snapshot-hash` is just a text file
<!-- Reviewable:start -->
[<img src="https://reviewable.io/review_button.png" height=40 alt="Review on Reviewable"/>](https://reviewable.io/reviews/servo/servo/6682)
<!-- Reviewable:end -->
Add memory profiling for the compositor task
Currently only the BufferMap is recorded, but a later change will also
measure the memory usage of the compositor tree.
<!-- Reviewable:start -->
[<img src="https://reviewable.io/review_button.png" height=40 alt="Review on Reviewable"/>](https://reviewable.io/reviews/servo/servo/6640)
<!-- Reviewable:end -->
Implement HSTS (preload-only)
Implement HSTS (preload-only) servo/servo#6105
* Downloads the HSTS preload list from the chromium repo (same as gecko), then convert it to a list appropriate for servo.
* Reads the preload list when creating a resource task, and implements STS for those domains.
Still todo:
* Read Strict-Transport-Security headers from servers and add details to the in-memory HSTS list. (note: this requires hyper or servo to implement an STS header struct. Hyper seems like the appropriate location, so I will create an issue/PR there soon). The work for this is nearly done with the exception of adding a new ControlMsg and the new header.
* Persist HSTS list to disk with known hosts (perhaps a different issue should be raised for this?)
<!-- Reviewable:start -->
[<img src="https://reviewable.io/review_button.png" height=40 alt="Review on Reviewable"/>](https://reviewable.io/reviews/servo/servo/6490)
<!-- Reviewable:end -->
