This changeset implements the beginnings of fine-grained measurement of
Servo's data structures.
- It adds a new `SizeOf` trait, which is used to measure the memory used
by heap data structures, and implements it for some std types: Box,
String, Option, Arc, Vec, and DList.
- It adds a new `MemoryReporter` trait which is used to report memory
measurements from other threads to the memory profiler. Reporters are
registered and unregistered with the memory profiler, and the memory
profiler makes measurement requests of reporters when necessary.
- It plumbs a MemoryProfilerChan through to the layout task so it can
register a memory reporter.
- It implements the `SizeOf` trait for `DisplayList` and associated
types, and adds a memory reporter that uses it.
The display list hits 14.77 MiB when viewing
tests/html/perf-rainbow.html, and 2.51 MiB when viewing the Guardians of
the Galaxy Wikipedia page from servo-static-suite. Example output:
0.29: display-list::http://www.reddit.com/
0.00: display-list::http://static.adzerk.net/reddit/ads.html?sr=-reddit.com,loggedout&bust2#http://www.reddit.com
0.00: display-list::http://www.reddit.com/static/createadframe.html
There are a number of FIXME comments indicating sub-optimal things. This
is a big enough change for now that doing them as follow-ups seems best.
`background-size` per CSS-BACKGROUNDS § 3.9.
Nearest neighbor interpolation is used for `crisp-edges`, like Firefox.
A note has been added that we could do better if we wanted to.
Multiple backgrounds are not yet supported.
Move culling of transparent display items to paint task rather than display list builder, so that hit testing detects mouse over on transparent background elements.
This patch will iterate through the DisplayList after the reflow
is done and print its elements (as also any sub-lists associated
to a child node stacking context).
It adds a new CLI parameter to trigger the function to dump the display list
to console (i.e. servo --debug dump-display-list url).
Using both display list and flow tree information is helpful to debug
rendering issues.
This will also update the rust-azure dependency
to point to the hash where we have a single Color type.
Just executed ref-tests and no regressions found.
This used to conflict with the util crate from the standard library, which
has long since been removed.
The import in layout has not been changed because of a conflict with the
util mod there.
through display list building.
The old `flow_origin` concept was ill-defined (sometimes the border box
plus the flow origin, sometimes including horizontal margins and
sometimes not, sometimes including relative position and sometimes not),
leading to brittleness and test failures. This commit reworks the logic
to always pass border box origins in during display list building.
This fixes the following warning:
display_list/mod.rs:735:20: 735:30 warning: use of deprecated item: Use `Int::zero()` or `Float::zero()`., #[warn(deprecated)] on by default
display_list/mod.rs:735 let zero = Zero::zero();
^~~~~~~~~~
`invert` is not yet supported.
Objects that get layers will not yet display outlines properly. This is
because our overflow calculation doesn't take styles into account and
because layers are always anchored to the top left of the border box.
Since fixing this is work that is not related to outline *per se* I'm
leaving that to a followup and making a note in the code.
This exposed some problems in our clipping logic, which was never
properly rewritten for the stacking context reform. The clipping code
worked in terms of a stack of clips, but the new stacking context code
has no concept of a stack of clip regions. Fixing that in turn exposed
some flaky/incorrect tests:
* `borders` had an incorrect reference image, as far as I can tell.
* `negative_margins` had some stray pixels, fixed by changing the text.
This adds the infrastructure necessary to support stacking contexts that
are not containing blocks for absolutely-positioned elements. Our
infrastructure did not support that before. This minor revamp actually
ended up simplifying the logic around display list building and
stacking-relative position computation for absolutely-positioned flows,
which was nice.
This patch is a first stab at implementing border-radius. It looks fine as long as
the border isn't an ellipse (that might not even parse yet), and the border-widths
around a border-radius are the same.
Here's a cool screenshot!
r? @pcwalton @SimonSapin
first-class.
This implements the scheme described here:
https://groups.google.com/forum/#!topic/mozilla.dev.servo/sZVPSfPVfkg
This commit changes Servo to generate one display list per stacking
context instead of one display list per layer. This is purely a
refactoring; there are no functional changes. Performance is essentially
the same as before. However, there should be numerous future benefits
that this is intended to allow for:
* It makes the code simpler to understand because the "new layer needed"
vs. "no new layer needed" code paths are more consolidated.
* It makes it easy to support CSS properties that did not fit into our
previous flat display list model (without unconditionally layerizing
them):
o `opacity` should be easy to support because the stacking context
provides the higher-level grouping of display items to which opacity
is to be applied.
o `transform` can be easily supported because the stacking context
provides a place to stash the transformation matrix. This has the side
benefit of nicely separating the transformation matrix from the
clipping regions.
* The `flatten` logic is now O(1) instead of O(n) and now only needs to
be invoked for pseudo-stacking contexts (right now: just floats),
instead of for every stacking context.
* Layers are now a proper tree instead of a flat list as far as layout
is concerned, bringing us closer to a production-quality
compositing/layers framework.
* This commit opens the door to incremental display list construction at
the level of stacking contexts.
Future performance improvements could come from optimizing allocation of
display list items, and, of course, incremental display list
construction.
I'm sad to say that this improved performance significantly. A lot of
this win is due to the Rust compiler not being smart about not zeroing
objects out if it doesn't need to.
