Here we add a new UserFontLoadState value, STATUS_LOAD_PENDING, which
represents the state just after a gfxUserFontEntry's url()-valued source
would being loading, except that we can't start the load due to being
on a Servo style worker thread. In that case, we defer the work of
initiating the load until just after the Servo traversal is finished.
URLs that can normally be loaded synchronously, such as data: URLs
and script-implemented protocols marked as synchronous, must be
handled asynchronously when encountered during Servo traversal, since
various main-thread only work (in FontFaceSet::SyncLoadFontData) must
happen. This is a user visible change from stock Gecko, but should
only happen when font metrics for a data: URL font are requested
due to ch/ex unit resolution when layout hasn't previously requested
the font load. Hopefully nobody relies on synchronous resolution of
ch/ex units with data: URLs.
We unfortunately also can't pick gfxUserFontEntry objects out of the
UserFontCache during Servo traversal, since validating the cache
entry involves doing content policy checking, which is not thread-safe
(due in part to taking strong references to nsIPrincipals).
Platform fonts and ArrayBuffer-backed DOM FontFace objects continue
to be handled synchronously.
The PostTraversalTask does not take a strong reference to the
gfxUserFontEntry object, since it is held on to by the DOM FontFace
object, which itself won't go away before the PostTraversalTask
is run.
MozReview-Commit-ID: J9ODLsusrNV
--HG--
extra : rebase_source : d3e3d1dc187cb252750b57bcecd0b1ed77a15a7c
The PostTraversalTask does not take a strong reference to the FontFaceSet
object, since as a DOM object, we can't call AddRef/Release on it
from the Servo style worker threads. The FontFaceSet is held on to
by the nsIDocument, and only cleared during cycle collection, which
should not occur between the font metrics request and the
PostTraversalTask running.
MozReview-Commit-ID: 5aKIg4DIQ4w
--HG--
extra : rebase_source : a7b63582b3610491fbcb4e2b931d55681e626513
The PostTraversalTask does not take a strong reference to the FontFace
object, since as a DOM object, we can't call AddRef/Release on it
from the Servo style worker threads. The FontFace objects that we
encounter are all held on to by the FontFaceSet, and none of the
work that we do during font loads should drop FontFace objects
from the FontFaceSet. (That only happens under
nsIDocument::FlushUserFontSet, which is only called on the
main thread.)
MozReview-Commit-ID: 5CdtGQYC9aL
--HG--
extra : rebase_source : 8ce5ff0c4ca306ccc240a86e7cacb588cc6c94c7
In a later patch, we'll want to queue up some tasks to run when the
Servo traversal is one, and the ServoStyleSet seems like the natural
place to store those tasks. We could probably find the ServoStyleSet
by chasing a bunch of pointers from the task-adding call sites, but
it seems simpler just to make it available directly.
MozReview-Commit-ID: AJoFZEoNaGm
--HG--
extra : rebase_source : 78389d72ba6dfb4301ba75bd39bbdc51d13cb4d5
Here we add a new UserFontLoadState value, STATUS_LOAD_PENDING, which
represents the state just after a gfxUserFontEntry's url()-valued source
would being loading, except that we can't start the load due to being
on a Servo style worker thread. In that case, we defer the work of
initiating the load until just after the Servo traversal is finished.
URLs that can normally be loaded synchronously, such as data: URLs
and script-implemented protocols marked as synchronous, must be
handled asynchronously when encountered during Servo traversal, since
various main-thread only work (in FontFaceSet::SyncLoadFontData) must
happen. This is a user visible change from stock Gecko, but should
only happen when font metrics for a data: URL font are requested
due to ch/ex unit resolution when layout hasn't previously requested
the font load. Hopefully nobody relies on synchronous resolution of
ch/ex units with data: URLs.
We unfortunately also can't pick gfxUserFontEntry objects out of the
UserFontCache during Servo traversal, since validating the cache
entry involves doing content policy checking, which is not thread-safe
(due in part to taking strong references to nsIPrincipals).
Platform fonts and ArrayBuffer-backed DOM FontFace objects continue
to be handled synchronously.
The PostTraversalTask does not take a strong reference to the
gfxUserFontEntry object, since it is held on to by the DOM FontFace
object, which itself won't go away before the PostTraversalTask
is run.
MozReview-Commit-ID: J9ODLsusrNV
--HG--
extra : rebase_source : 1651e2917bd31b87fc1c1be94b0eced1273df86a
The PostTraversalTask does not take a strong reference to the FontFaceSet
object, since as a DOM object, we can't call AddRef/Release on it
from the Servo style worker threads. The FontFaceSet is held on to
by the nsIDocument, and only cleared during cycle collection, which
should not occur between the font metrics request and the
PostTraversalTask running.
MozReview-Commit-ID: 5aKIg4DIQ4w
--HG--
extra : rebase_source : 50ca5f2d8af42cfbeb6808aa755a4670bdf293bd
The PostTraversalTask does not take a strong reference to the FontFace
object, since as a DOM object, we can't call AddRef/Release on it
from the Servo style worker threads. The FontFace objects that we
encounter are all held on to by the FontFaceSet, and none of the
work that we do during font loads should drop FontFace objects
from the FontFaceSet. (That only happens under
nsIDocument::FlushUserFontSet, which is only called on the
main thread.)
MozReview-Commit-ID: 5CdtGQYC9aL
--HG--
extra : rebase_source : 8e4f21a12ae5ca81017293d5b06f1551e307ea5d
In a later patch, we'll want to queue up some tasks to run when the
Servo traversal is one, and the ServoStyleSet seems like the natural
place to store those tasks. We could probably find the ServoStyleSet
by chasing a bunch of pointers from the task-adding call sites, but
it seems simpler just to make it available directly.
MozReview-Commit-ID: AJoFZEoNaGm
--HG--
extra : rebase_source : d8afe0c2cdf1ea4b1681c6e57aea48c9efddea00
In our current setup, in which links with an href attribute always match either
:link or :visited, no matter whether that attribute's value is a valid URI,
changes to the attribute always put the element into either the "match nothing"
state or the "match :link" state, via calls to Link::ResetLinkState.
The only thing FlushPendingLinkUpdates is needed for is (lazily, in case it
turns out to not be needed because the element got removed from the DOM anyway)
registering a history observer to switch the link state to :visited as needed.
This means that selector matching consumers that would never expose :visited
state to start with don't need to worry about calling FlushPendingLinkUpdates.
Fixed backgrounds were not being clipped correctly when rendered using
inactive layers.
MozReview-Commit-ID: 3v8tajr3MoB
--HG--
extra : rebase_source : 25329cb9efb998c64924916d4ea496a439688c8d
Continue to allow non-multiprocessCompatible extensions in automation.
There are a ton of places that would need to be changed, many of which
will be changing soon anyway with the non-webextensions change in 57
so this is mostly the expedient route to keeping the tree green.
MozReview-Commit-ID: EZZoDVdhLfy
--HG--
extra : rebase_source : f83472bc1c88dd0deadbe485d9002499027ff07f
Currently, we only correctly support remote layer trees for frameloaders that
use the same layer manager as their document. Since we need to be able to host
remote <browser> content in popup widgets for remote WebExtensions, we need to
tie the frameloaders to the layer manager of their host element, rather than
the root layer manager for the document.
MozReview-Commit-ID: 4RCsamFBiQw
--HG--
extra : rebase_source : 86bca4ae1c012ff1bb84a9ad796be311cfe580f6
extra : histedit_source : 19577d69430adc8cb38c195f13db2c6de6605c4c
Previously the harness would check for navigator.plugins[Test Plug-in] in the browser window, but that is now blocked from working. Instead we can simply use an existing getTestPlugin function that enumerates the plugin through nsIPluginHost
MozReview-Commit-ID: Kp48u5iFkSa
To mitigate the delay that a new content process startup might cause, when the
browser reaches a non-busy state we attempt to prelaunch an empty content process
in the background that can be grabbed and used the next time we need one. This patch
enables the preallocated process manager by default and attempts to fix all the
issues that prevented us doing this sooner.
That function has a special path for when there's no parent reflow input, and if
there's a parent reflow input there should always be a parent frame.
MozReview-Commit-ID: EYh0aE4ozBg
By passing the startTime as a TimeDuration we are able to represent times in the
distant past (and with the same range as we can represent on the main thread so
that if we do encounter range errors in future, they should not differ between
the main thread and the compositor).
This patch includes a crashtest. I have verified that, without the code changes
included in this patch, this crashtest fails on debug builds on OSX.
MozReview-Commit-ID: EDuKLzfEC0K
--HG--
extra : rebase_source : 1883080fdfac8c33f70698145f21e67cbdfdd4f2
Cameron McCormack
Iris Hsiao
Carsten "Tomcat" Book
Hiroyuki Ikezoe
Boris Zbarsky
Wes Kocher
Jamie Nicol
Andrew Swan
Kris Maglione
Wei-Cheng Pan
Botond Ballo
Jonathan Watt
Benjamin Smedberg
Gabor Krizsanits
Emilio Cobos Álvarez
Brad Werth
Thomas Nguyen
Brian Birtles
Robert Longson