We optimize the surface contents in CanvasImageCache for the particular
DrawTarget backend used by the canvas. If different backends are
intermixed, then the cache may not be properly optimized (e.g. Skia and
Direct2D). This can have performance implications.
This patch adds the backend type to the cache entries, so that we ignore
cache entries created with a different backend type.
Differential Revision: https://phabricator.services.mozilla.com/D159388
The remaining methods in ns[T]StringObsolete are all find+replace methods for
nsTSubstring. These were migrated in a similar way to the find methods, and
partially updated to avoid using methods from nsStringObsolete.cpp.
This change removes the ns[T]StringObsolete.cpp files completely, as they are
no longer necessary.
Differential Revision: https://phabricator.services.mozilla.com/D148301
Add a pref for MouseEvent.region since that wasn't un-exposed. No other
browser supports it so we can probably safely remove it, but just in
case.
Differential Revision: https://phabricator.services.mozilla.com/D152274
gfxFontCache acquires and releases its mutex during various operations.
In order to keep the state internally consistent, we should only release
the lock after the full operation is complete. This involves moving the
deletion of gfxFont to outside the lock via a temporary discard array.
The expiration state should not be protected by the gfxFont's mutex
since we don't hold it during most operations. Instead we should hold
gfxFontCache's mutex because then we can guarantee the operation is
atomic, particularly when a worker wants a font, and the main thread is
aging the generations.
When a font is returned from gfxFontCache, we now return it already
removed from the tracker, and with its refcount updated. This avoids any
potential races between the expiration timer and a worker accessing the
font, as well as simplying the callers so they don't need to be aware of
addref-ing manually in case the result is to be discarded (so that it
gets readded to the tracker).
Differential Revision: https://phabricator.services.mozilla.com/D151821
gfxFontCache acquires and releases its mutex during various operations.
In order to keep the state internally consistent, we should only release
the lock after the full operation is complete. This involves moving the
deletion of gfxFont to outside the lock via a temporary discard array.
The expiration state should not be protected by the gfxFont's mutex
since we don't hold it during most operations. Instead we should hold
gfxFontCache's mutex because then we can guarantee the operation is
atomic, particularly when a worker wants a font, and the main thread is
aging the generations.
When a font is returned from gfxFontCache, we now return it already
removed from the tracker, and with its refcount updated. This avoids any
potential races between the expiration timer and a worker accessing the
font, as well as simplying the callers so they don't need to be aware of
addref-ing manually in case the result is to be discarded (so that it
gets readded to the tracker).
Differential Revision: https://phabricator.services.mozilla.com/D151821
All its members are optional, so we can just use it as a plain struct
rather than Maybe<> all around, which simplifies the code and prevents
silly bugs like bug 1779592.
Mostly automatic via:
rg -l 'SVGImageContext' . | xargs sed -i 's/Maybe<SVGImageContext>/SVGImageContext/g'
With trivial build fixes.
Not intended to change behavior.
Differential Revision: https://phabricator.services.mozilla.com/D151846
Per spec, the value of fontKerning should be an enum, not a string, but currently we handle
all the keyword-valued canvas attributes in this way. We may want to convert them to enums
(which will mean that unrecognized values throw an error instead of being ignored), but I
think that should be done for all the attributes together as a separate bug. For now, using
a string here provides consistency with the rest of the canvas APIs.
Note that Blink's current implementation and the existing WPT tests have some problems:
they treat the values of fontKerning as case-insensitive, which is wrong. I have filed
https://bugs.chromium.org/p/chromium/issues/detail?id=1343333 about this.
Differential Revision: https://phabricator.services.mozilla.com/D151755
The remaining methods in ns[T]StringObsolete are all find+replace methods for
nsTSubstring. These were migrated in a similar way to the find methods, and
partially updated to avoid using methods from nsStringObsolete.cpp.
This change removes the ns[T]StringObsolete.cpp files completely, as they are
no longer necessary.
Differential Revision: https://phabricator.services.mozilla.com/D148301
Add a dom/base/rust crate called just "dom" where we can share these.
Most of the changes are automatic:
s/mozilla::EventStates/mozilla::dom::ElementState/
s/EventStates/ElementState/
s/NS_EVENT_STATE_/ElementState::/
s/NS_DOCUMENT_STATE_/DocumentState::/
And so on. This requires a new cbindgen version to avoid ugly casts for
large shifts.
Differential Revision: https://phabricator.services.mozilla.com/D148537
None of the consumer need to mutate styles, and this saves some ugly
const_casting on the next patch.
Doesn't change behavior.
Differential Revision: https://phabricator.services.mozilla.com/D147555
Bug 1719886 tried to simplify this code by using Rect::Intersect, but that had the
side-effect of it trying to detect if either width OR height was <= 0, and if so,
setting both width AND height to 0. Pattern dimensions need to be limited independently
here for correct behavior, so we just default to inlining the code for each dimension
without the incorrect final behavior of zeroing the rectangle.
Differential Revision: https://phabricator.services.mozilla.com/D146144
Since we're going to be keeping around empty cache pages, which can further
increase memory pressure, it would be nice if we respond to memory pressure
events by evicting the empty cache pages.
Differential Revision: https://phabricator.services.mozilla.com/D145499
In general, the canvas text-drawing code makes two passes over the string, one to measure it and the second to draw.
Each time, it has to find individual direction runs and create a separate gfxTextRun for each.
However, in the (common) case of a simple unidirectional string, we're only going to have a single run,
so it's wasteful to construct it twice; the CanvasBidiProcessor can just re-use the run from the measurement
pass when it needs to draw.
In particular, this will apply to pdf.js drawing, where each glyph is handled in a separate fillText call.
Many other uses of canvas text should also benefit somewhat.
Differential Revision: https://phabricator.services.mozilla.com/D145426
This enables an offscreen-canvas Worker to do text rendering, but only with system-installed fonts
because we don't yet have a FontFaceSet available in the Worker. (That's bug 1758946.)
Differential Revision: https://phabricator.services.mozilla.com/D144188
These need to be atomic so that they can be accessed by canvas code when running on a worker thread
without calling main-thread-only Preferences APIs.
Depends on D144185
Differential Revision: https://phabricator.services.mozilla.com/D144186
We don't need to null-check the mCanvasElement and mDocShell members in these methods,
and bail out if neither is provided, because in that case the following GetPresShell() call
will return null and we'll bail out there anyhow.
Removing these extra checks will simplify the changes involved in enabling the canvas code
to be used in a Worker context, where it is not connected to a document.
Depends on D144184
Differential Revision: https://phabricator.services.mozilla.com/D144185
Nothing uses this member, so remove it to make it clear we don't need to provide it
when setting up a CanvasRenderingContext2D.
Differential Revision: https://phabricator.services.mozilla.com/D144184
This requires updating a couple of our old dom/canvas mochitests that assume the size element
of the font attribute will be serialized as specified, which conflicts with the current spec
requirement.
On the other hand, this also resolves a few current WPT test failures that are specifically
testing the behavior described in the spec (e.g. a size specified as percentage gets serialized
as the computed px value).
Differential Revision: https://phabricator.services.mozilla.com/D144812
Canvas 2D needs paint callbacks from the OffscreenCanvas display
pipeline so that it knows when to reinvalidate. They were disabled which
meant only the first frame was rendered.
Differential Revision: https://phabricator.services.mozilla.com/D144006
This adds OnEvent hooks to DrawTargetWebgl for various events so that
profile counters may be maintained each frame. These profile counters
try to determine if a software fallback happened or an uncacheable event
occurs that requires either uploading data to the GPU or reading back
from the GPU, events which can cause substantial slowdown if they happen
repeatedly even without an explicit fallback to software rasterization.
When it is determined a threshold has been reached (as controlled by
some prefs), RequiresRefresh() in PersistentBufferProvider is used to
signal that we should recreate the PersistentBufferProvider of a different
type and thus disable acceleration.
Differential Revision: https://phabricator.services.mozilla.com/D142646
This implements a new DrawTarget API call, DrawShadow. This can be used to draw
the shadow of a given Path and Pattern. Unlike DrawSurfaceWithShadow, this only
draws the shadow to simplify the resulting implementation. DrawTarget provides
a default implementation that will draw a blurred shadow to a transient surface
before handing it off to DrawSurfaceWithShadow, allowing existing DrawTarget
implementations to function without having to implement the new API at all.
Within DrawTargetWebgl, DrawShadow allows direct caching of the shadow in the
existing PathCache mechanism, unlike DrawSurfaceWithShadow which would usually
be called with different transient surfaces every time, even if the input
patterns actually matched.
CanvasRenderingContext2D is modified to call DrawShadow when only a shadow
needs to be applied without any filter. AdjustedTarget is modified to wrap
the necessary DrawTarget draw calls so they can be dispatched appropriately.
Since a lot of redundant shadow parameters are shared between DrawShadow and
DrawSurfaceWithShadow, and since a lot of consumers of these functions need
to also pass around these parameters, this also bundles these into a new
ShadowOptions structure to make this easier.
Differential Revision: https://phabricator.services.mozilla.com/D142026
This patch ensures we return an empty surface as per the spec when
the OffscreenCanvas passed into CreatePattern and DrawImage has no
context bound to it.
Differential Revision: https://phabricator.services.mozilla.com/D139127
Most of the support for presenting a WebGLFramebuffer to a swap chain existed as part of the
mechanism for opaque WebXR framebuffer support. However, such "opaque" framebuffer are meant
to be opaque in the sense that their attachments can't be inspected or changed, which does
not provide the requisite level of control for efficiently implementing Canvas2D snapshots.
To this end, the existing Present mechanism is slightly extended to allow presenting to the
swap chain already present in WebGLFramebuffer without the existence of a corresponding
MozFramebuffer.
This also fixes a bug in that AsWebgl() was no longer being utilized in CanvasRenderer, such
that a new mechanism that routed GetFrontBuffer() was needed to fix the code rot.
There are also some efforts to remove a couple redundant copies I noticed in profiles along
the way.
Differential Revision: https://phabricator.services.mozilla.com/D138119
Most of the support for presenting a WebGLFramebuffer to a swap chain existed as part of the
mechanism for opaque WebXR framebuffer support. However, such "opaque" framebuffer are meant
to be opaque in the sense that their attachments can't be inspected or changed, which does
not provide the requisite level of control for efficiently implementing Canvas2D snapshots.
To this end, the existing Present mechanism is slightly extended to allow presenting to the
swap chain already present in WebGLFramebuffer without the existence of a corresponding
MozFramebuffer.
This also fixes a bug in that AsWebgl() was no longer being utilized in CanvasRenderer, such
that a new mechanism that routed GetFrontBuffer() was needed to fix the code rot.
There are also some efforts to remove a couple redundant copies I noticed in profiles along
the way.
Differential Revision: https://phabricator.services.mozilla.com/D138119
This patch adds a partial OffscreenCanvasRenderingContext2D
implementation. It is missing anything text and UI related, including
CanvasFilters, CanvasUserInterface, CanvasText, CanvasTextDrawingStyles,
and CanvasHitRegions.
Differential Revision: https://phabricator.services.mozilla.com/D135354
This patch is a non-functional change. It overloads some WebIDL methods
of CanvasRenderingContext2D to account for the difference between the
main thread and worker prototypes.
Differential Revision: https://phabricator.services.mozilla.com/D135353
This patch is a non-functional change which allows the shutdown observer
to be overridden (since workers have a different mechanism). It also
fixes a static counter to be atomic, and adds mOffscreenCanvas to the
cycle collection tracking.
Differential Revision: https://phabricator.services.mozilla.com/D135352
