Commit graph

513 commits

Author SHA1 Message Date
Nicholas Nethercote
7429b90e02 Wire up the JS engine's memory reporting.
SpiderMonkey provides an extremely fine-grained breakdown of memory
usage, but for Servo we aggregate the measurements into a small number
of coarse buckets, which seems appropriate for the current level of
detail provided by Servo's memory profiler. Sample output:
```
|   10.99 MiB -- pages
|       7.75 MiB -- url(http://html5demos.com/worker)
|          4.63 MiB -- js
|             2.00 MiB -- gc-heap
|                0.94 MiB -- decommitted
|                0.92 MiB -- used
|                0.09 MiB -- unused
|                0.05 MiB -- admin
|             1.44 MiB -- malloc-heap
|             1.19 MiB -- non-heap
|          [...]
|       3.24 MiB -- url(http://html5demos.com/js/worker-cruncher.js)
|          3.24 MiB -- js
|             1.17 MiB -- malloc-heap
|             1.06 MiB -- non-heap
|             1.00 MiB -- gc-heap
|                0.69 MiB -- used
|                0.19 MiB -- decommitted
|                0.09 MiB -- unused
|                0.03 MiB -- admin
```
Most of the changes are plumbing to get the script and worker tasks
communicating with the memory profiler task.
2015-07-15 21:58:20 -07:00
bors-servo
9af229b830 Auto merge of #6591 - servo:rustup_2015-07-10, r=larsbergstrom
Upgrade to rustc 1.3.0-dev (fddfd089b 2015-07-10)

Depends on https://github.com/jgraham/webdriver-rust/pull/12.

Fixes #6020.

<!-- Reviewable:start -->
[<img src="https://reviewable.io/review_button.png" height=40 alt="Review on Reviewable"/>](https://reviewable.io/reviews/servo/servo/6591)
<!-- Reviewable:end -->
2015-07-14 13:40:22 -06:00
Martin Robinson
efdf2cd601 Update to latest rust-layers, azure, and skia
GLRasterizationContext is now responsible for doing GPU rasterization.
It can coexist with its target NativeSurface, so we don't have to
continually recreate NativeSurfaces when doing GPU rasterization.
2015-07-14 17:48:50 -07:00
Patrick Walton
c2195c490d compositing: Use the pre-existing script channel in the same-domain
iframe case instead of creating a new one.

Fixes test timeouts.
2015-07-14 15:27:36 -07:00
Patrick Walton
e5b1ec4078 script: Split Pipeline::create into chrome process and content
process parts.

This will make it easier to adapt to IPC.

The trickiest part here was to make script tasks spawn new layout tasks
directly instead of having the pipeline do it for them. The latter
approach will not work in multiprocess mode, because layout and script
must run in the same address space and the pipeline cannot inject tasks
into another process.
2015-07-14 15:09:16 -07:00
Patrick Walton
9ce65c08a5 layout: Make the LayoutControlChan use IPC. 2015-07-14 13:45:52 -07:00
Patrick Walton
e841065351 compositing: Move messages that go over the ScriptListener to go over
an IPC channel instead.

Because this used a boxed trait object to invoke messages across a
process boundary, and boxed trait objects are not supported across IPC,
we spawn a helper thread inside the compositor to perform the marshaling
for us.
2015-07-14 12:12:19 -07:00
Martin Robinson
e115c3d3c4 Move LayerBuffer cache to the compositor
Now that NativeDisplay can be shared between the compositor and the
paint task, we can move the LayerBuffer cache to the compositor. This
allows surfaces to be potentially reused between different paint tasks
and will eventually allow OpenGL contexts to be preserved between
instances of GL rasterization.
2015-07-08 08:05:11 -07:00
Matt Brubeck
8570b22677 Remove layers dependency from util 2015-07-01 09:40:32 -07:00
bors-servo
4674afe846 Auto merge of #6525 - mrobinson:simplify-display, r=glennw
Update to latest rust-layers

The compositing context, painting context and display metadata have all
been collapsed into a single NativeDisplay class.

<!-- Reviewable:start -->
[<img src="https://reviewable.io/review_button.png" height=40 alt="Review on Reviewable"/>](https://reviewable.io/reviews/servo/servo/6525)
<!-- Reviewable:end -->
2015-06-30 17:02:26 -06:00
Martin Robinson
0f27bd6c4b Update to latest rust-layers
The compositing context, painting context and display metadata have all
been collapsed into a single NativeDisplay class.
2015-06-30 21:58:54 -07:00
Jesse Ruderman
1ff4fe02d9 Add keyboard shortcuts to glutin browser 2015-06-27 03:06:18 -07:00
Ms2ger
73a7e92bfd Update to rustc 2d0cbf3e3e25e092bd9e4c94d08e446b680869f0. 2015-06-25 20:02:36 +02:00
Glenn Watson
39ddbbb0e1 Implement enough of 3d transforms spec to run the CSS FPS demo. 2015-06-23 14:10:44 +10:00
ecoal95
8cbfb3482c Use euclid from crates.io 2015-06-19 00:04:24 +02:00
James Graham
22c06307b8 Implement timeout for url load from WebDriver. 2015-06-17 00:25:23 +01:00
Simon Sapin
a80a12e41e Use cgl and gleam from crates.io 2015-06-15 23:00:33 +02:00
Corey Farwell
5c408d2be9 rust-geom API changes
https://github.com/servo/rust-geom/pull/81
2015-06-13 12:28:21 -07:00
Jack Moffitt
b5ebbaefe3 Clean up dependencies. 2015-06-07 12:14:38 -06:00
Eduard Burtescu
00e8e5f77d Use the correct log crate and setup env_logger in main. 2015-06-06 23:20:40 +03:00
Mukilan Thiyagarajan
6413526440 Simplify handle_load_start_msg function 2015-06-05 19:43:26 +05:30
Corey Farwell
a14c745a6a Bump rust-url version: 0.2.16 → 0.2.35 2015-06-04 16:45:15 -04:00
bors-servo
b0f3417cff Auto merge of #6219 - zmike:favicons_and_stringlOWWW-MY-HEAD, r=pcwalton
Getting these down to the embedding API level required that I redo the bindings generator again, so this is more commits than anticipated.

@mbrubeck @Manishearth @pcwalton but NOT @larsbergstrom so don't even look at this.

<!-- Reviewable:start -->
[<img src="https://reviewable.io/review_button.png" height=40 alt="Review on Reviewable"/>](https://reviewable.io/reviews/servo/servo/6219)
<!-- Reviewable:end -->
2015-06-03 00:27:41 -05:00
Mike Blumenkrantz
ffa2093012 add window method for notifying when the <head> tag has been parsed 2015-06-02 22:16:28 -04:00
Mike Blumenkrantz
12f20f1ecc add handling for favicon link elements
fixes #6166
2015-06-02 22:16:25 -04:00
bors-servo
0b7886b15f Auto merge of #6161 - jgraham:webdriver_stability, r=glennw
This fixes various race conditions that affect test execution when using the servodriver product. It doesn't yet do enough to make servodriver a viable alternative to the normal servo test executor.

<!-- Reviewable:start -->
[<img src="https://reviewable.io/review_button.png" height=40 alt="Review on Reviewable"/>](https://reviewable.io/reviews/servo/servo/6161)
<!-- Reviewable:end -->
2015-06-02 21:11:44 -05:00
bors-servo
f6fe195334 Auto merge of #6230 - frewsxcv:getters-get, r=nox
Part of #6224

I certainly didn't remove all of them; I avoided `unsafe` areas and also `components/script`

<!-- Reviewable:start -->
[<img src="https://reviewable.io/review_button.png" height=40 alt="Review on Reviewable"/>](https://reviewable.io/reviews/servo/servo/6230)
<!-- Reviewable:end -->
2015-06-02 08:17:03 -05:00
Corey Farwell
435e551753 Remove get_ prefix on getters
Part of #6224

I certainly didn't remove all of them; I avoided `unsafe` areas and also `components/script`
2015-06-02 08:54:44 -04:00
James Graham
aa0f7a7a12 Improve the stability of tests run using WebDriver.
* Wait for the correct pipeline to return a LoadComplete message
  before signalling a load is complete, and ensure that the root
  pipeline is the one corresponding to the active document of the top
  level browsing context, even if this has not yet painted.

* Ensure that TakeScreenshot operates on the correct pipeline

* Reset the screenshot ready flag whenever we decide that we are ready
  to take a screenshot.
2015-06-02 11:13:18 +01:00
James Graham
2b396b2485 Allow Window implementations to indicate that they don't support a clipboard.
This is important for the SERVO_HEADLESS configuration, because
creating a clipboard on linux creates an X context which then causes
reftest instability.
2015-06-02 08:57:09 +01:00
James Graham
49f1b13ad9 Add support for switching frames with the webdriver API.
This moves webdriver_traits into msg to avoid a circular dependency.
2015-06-01 23:18:55 +01:00
Nicholas Nethercote
a21f6c407c Measure paint task buffer maps in the memory profiler.
Example output from the memory profiler:

```
|       1.04 MiB -- url(http://en.wikipedia.org/wiki/Main_Page)
|          0.26 MiB -- display-list
|          0.78 MiB -- paint-task       # new output line
|             0.78 MiB -- buffer-map    # new output line
```

The buffer maps aren't huge, but they're worth measuring, and it's good
to get the memory profiler plumbing into PaintTask.
2015-05-27 18:27:18 -07:00
Mike Blumenkrantz
fcf449517d handle case of deleted pipeline when preparing to send load_end constellation msg 2015-05-27 14:45:56 -04:00
Mike Blumenkrantz
64ff2c4a54 handle early LoadURL messages without panicking by sending initial url load
fixes #6126
2015-05-27 14:45:54 -04:00
Mike Blumenkrantz
68fc51205f add error enum for chromium net errors, create window method for error propagation 2015-05-27 14:45:50 -04:00
Mike Blumenkrantz
0d46a3b89a add navigation state data to LoadComplete messages, finish cef load state cbs 2015-05-27 14:45:48 -04:00
Mike Blumenkrantz
3481c752cd add CompositorMsg::LoadStart, implement cef_load_handler::on_loading_state_change()
only adds the loading:true callback this time...
2015-05-27 14:45:47 -04:00
Mike Blumenkrantz
92d44fb36e send ChangeUrl message from constellation on init 2015-05-27 14:45:41 -04:00
Mike Blumenkrantz
cf388cd909 call compositor.window.set_page_url() on WindowEvent::LoadUrl 2015-05-27 14:45:41 -04:00
Mike Blumenkrantz
fb72a64833 pass width and height information through to cef paint callback 2015-05-21 21:15:15 -04:00
Mike Blumenkrantz
9f7aacdacf add check for self.context existence in composite_specific_target() 2015-05-21 21:15:14 -04:00
bors-servo
ad53e95080 Auto merge of #6083 - ecoal95:layerize-canvas, r=pcwalton
I've done a bit of job to get this done. Right now readback is still used, but we have a `LayerId` -> `CanvasRenderer` map on the paint task, that we can use to get rid of that.

I'd want review, to see if this is a good approach (I know it's not the initial `CanvasId` -> renderer approach, but it's pretty similar, since a canvas involves a `PaintLayer`).

I had to do a bit of refactoring to avoid cyclic dependencies between canvas and gfx. I'd want you to review them too.

It's mergeable and doesn't break any tests :P

Some of my main concerns:
* Does the canvas render really need to be behind an `Arc<Mutex<T>>`?
* I can't clone a `NativeSurface` right now (that's why the `SendNativeSurface()` msg is unimplemented in the WebGL task). It should be easy to add that to rust-layers, supposing the caller is responsible to mark it as non-leaking, any reason to not do it?

cc @jdm @pcwalton

<!-- Reviewable:start -->
[<img src="https://reviewable.io/review_button.png" height=40 alt="Review on Reviewable"/>](https://reviewable.io/reviews/servo/servo/6083)
<!-- Reviewable:end -->
2015-05-20 15:42:06 -05:00
Patrick Walton
0098d9e9e8 compositing: Support multiple events per frame.
Improves scrolling performance on Mac.
2015-05-20 10:53:48 -07:00
ecoal95
3350522306 Layerize canvas
Note that this keeps using readback right now, `NativeSurface` painting
will be implemented soon.

Also see https://github.com/servo/servo/issues/6142
2015-05-20 19:10:50 +02:00
bors-servo
c51e9f0455 Auto merge of #6131 - glennw:jquery-exit-fix, r=jdm
This fixes a hang found while testing the jQuery test suite.

<!-- Reviewable:start -->
[<img src="https://reviewable.io/review_button.png" height=40 alt="Review on Reviewable"/>](https://reviewable.io/reviews/servo/servo/6131)
<!-- Reviewable:end -->
2015-05-19 21:44:45 -05:00
Patrick Walton
6a197719b3 compositing: Implement display ports and avoid creating display lists
for items outside it.

This improves Servo's performance on large pages.
2015-05-19 16:53:51 -07:00
Glenn Watson
35a570ab66 Fix several hangs / panics during pipeline cleanup of in progress loads.
This fixes a hang found while testing the jQuery test suite.
2015-05-19 16:14:25 +10:00
Ms2ger
0067cb2101 Remove some clones. 2015-05-15 17:52:53 +02:00
James Graham
28ac0abf6a Make WebDriver Get() command wait on pages loading before returning.
This makes using WebDriver significantly less racy. Also
refactors the message structure a little
2015-05-14 10:59:10 +01:00
Glenn Watson
2b3737d34e Various fixes for cleaning up iframes, compositor layers, pipelines and threads.
This allows most of the jquery test suite to run without exhausting thread resources.
2015-05-14 11:42:54 +10:00