b51ed22a3f
Bug 1284721 - Flush CanvasRenderingContext2D's DrawTarget before handing it off to the BufferProvider. r=lsalzman
2016-07-08 19:26:48 +02:00
ff42b1930a
Bug 1285207 - Don't let canvas have a DrawTarget without a BufferProvider and keep them in sync. r=lsalzman
2016-07-08 19:26:44 +02:00
0617cf9891
Bug 1284721 - Pref off PersistentBufferProviderShared due to various regressions. r=lsalzman
2016-07-08 19:26:42 +02:00
34daf6679c
Bug 1284632 - Move connecting.png and connecting@2x.png to shared/. r=dao
...
--HG--
rename : browser/themes/windows/tabbrowser/connecting.png => browser/themes/shared/tabbrowser/connecting.png
rename : browser/themes/windows/tabbrowser/connecting@2x.png => browser/themes/shared/tabbrowser/connecting@2x.png
2016-07-08 13:16:06 +02:00
37cacb6747
Merge mozilla-central to fx-team
2016-07-08 12:17:32 +02:00
838f21f656
merge mozilla-inbound to mozilla-central a=merge
2016-07-08 11:58:46 +02:00
75e7b5bab2
merge fx-team to mozilla-central a=merge
2016-07-08 11:57:15 +02:00
91979eebd5
Backed out changeset 68be1552a5ee (bug 1275546) for widespread failures in browser_css_angle.js
2016-07-08 11:25:08 +02:00
7220b22c54
Bug 1285409. Store out of flow data for frames we are going to descent into anyways (despite no intersection with the dirty rect) so we have the correct clip for them. r=mattwoodrow
...
If the frame already has the NS_FRAME_FORCE_DISPLAY_LIST_DESCEND_INTO bit we are going to descend into it anyways, so we may as well store the oof data so we can have the correct clip.
Sometimes we add the bit after this test though, so it doesn't allow us to have the proper clip in all cases.
2016-07-08 00:25:18 -05:00
1ed92648d1
Bug 1257037 part 15. Get rid of the wantsXrays configuration annotation. Just set bindings to want Xrays if they're exposed in Window. r=khuey
2016-07-08 00:37:55 -04:00
8c00d2132b
Bug 1257037 part 14. Remove the now-unused skipGen bits from binding code. r=khuey
2016-07-08 00:37:55 -04:00
d959247e48
Bug 1257037 part 13. Get rid of the concept of worker vs non-worker descriptors. r=khuey
2016-07-08 00:37:55 -04:00
4813a875a7
Bug 1257037 part 12. Simplify CGRegisterWorkerBindings now that there are no more worker descriptors. r=khuey
2016-07-08 00:37:55 -04:00
f3fde63185
Bug 1257037 part 11. Get rid of getRelevantProviders, since all the places that use it have a Configuration. r=khuey
2016-07-08 00:37:55 -04:00
584c6019b6
Bug 1257037 part 10. Store only one descriptor per interface name in the configuration. r=khuey
2016-07-08 00:37:54 -04:00
67d015d59c
Bug 1257037 part 9. Get rid of the concept of mainthread callbacks or worker callbacks; just codegen all the callbacks. r=khuey
2016-07-08 00:37:54 -04:00
ca29b3c56a
Bug 1257037 part 8. Remove Configuration.getDescriptorProvider. r=khuey
2016-07-08 00:37:54 -04:00
f719b4d8e4
Bug 1257037 part 7. Make the binding Configuration a non-workers DescriptorProvider. r=khuey
2016-07-08 00:37:54 -04:00
fc26f41e40
Bug 1257037 part 6. Simplify the toBindingNamespace function now that there are no worker bindings. r=khuey
2016-07-08 00:37:54 -04:00
e36bd30afb
Bug 1257037 part 5. Remove remaining checks for worker descriptors from Codegen.py. r=khuey
2016-07-08 00:37:54 -04:00
c65c962762
Bug 1257037 part 4. Remove the worker-only XXX comments from CGWrapGlobalMethod, since we never have worker descriptors for globals anyway. r=khuey
2016-07-08 00:37:54 -04:00
69caa834f3
Bug 1257037 part 3. Remove filteredSignatures, since it's a no-op unless we're dealing with a worker descriptor. r=khuey
2016-07-08 00:37:54 -04:00
1dbb14fbc5
Bug 1257037 part 2. Remove isMaybeExposedIn, since it always returns true now. r=khuey
2016-07-08 00:37:53 -04:00
69f3736864
Bug 1257037 part 1. Remove the worker-descriptor-related complexity in getControllingCondition, since workerDesc.workers never tests true anymore. r=khuey
2016-07-08 00:37:53 -04:00
32a933ec91
Bug 1283344 - disable dom/filesystem/tests/test_worker_basic.html for being number one with a bullet on the Orange Factor hit parade
2016-07-07 21:27:25 -07:00
8345db87d2
Bug 1282150 followup: placate the rooting analysis in SimpleGlobalObject::Create so we can reopen the CLOSED TREE
2016-07-07 22:24:37 -04:00
64abd594e5
Backed out 6 changesets (bug 1285036) for bustage on a CLOSED TREE.
...
Backed out changeset f7174077bd92 (bug 1285036)
Backed out changeset 4ae182434a6a (bug 1285036)
Backed out changeset 0390a7afb12d (bug 1285036)
Backed out changeset b93efbaf1d7a (bug 1285036)
Backed out changeset 3036026722de (bug 1285036)
Backed out changeset 8e94b77f805b (bug 1285036)
2016-07-07 21:57:49 -04:00
fbdd07f3b5
Bug 1283019. Only post a schedule composite if the compositor loop exists. r=dvander
2016-07-07 17:52:58 -07:00
b339d21385
Bug 1285239 - Scrolling can randomly get interrupted and show overscroll effect r=botond
2016-07-08 09:27:52 -07:00
2f30d77482
Bug 1279665 - skip aria-owns deferred accessible creation for XUL elements, r=yzen
2016-07-08 11:56:30 -04:00
ae02c9f568
Bug 1283882 - Update auto-generated bindings; r=me
2016-07-08 11:39:10 -04:00
ff2fd08d62
Bug 1283882 - Implement LOCATION_EVENT as native call; r=snorp
...
Convert LOCATION_EVENT in GeckoEvent to a native method call in
GeckoAppShell.
2016-07-08 11:39:10 -04:00
976590c3b2
Bug 1283882 - Move location listener to GeckoAppShell; r=snorp
...
Move LocationListener implementation from GeckoApp to GeckoAppShell
because the LocationListener consumer is in GeckoAppShell.
2016-07-08 11:39:09 -04:00
11c94a2c71
Bug 1283844 - Update auto-generated bindings; r=me
2016-07-08 11:39:09 -04:00
e01ee0921d
Bug 1283844 - Implement SENSOR_EVENT as native call; r=snorp
...
Remove SENSOR_EVENT from GeckoEvent and implement it as a native method
in GeckoAppShell that is invoked by the sensor event listener in
GeckoAppShell.
2016-07-08 11:39:09 -04:00
2b7ff3f3e5
Bug 1283844 - Implement sensor event listener in GeckoAppShell; r=snorp
...
Implement SensorEventListener directly in GeckoAppShell instead of
indirectly through GeckoInterface and GeckoApp, because the
SensorEventListener consumer is in GeckoAppShell.
2016-07-08 11:39:09 -04:00
a877b2bc01
Bug 1278265 - Give each JS shell thread its own SPS pseudo stack to avoid races. r=sfink
2016-07-08 17:31:30 +02:00
cae841f134
Bug 1281759 - Try to work around mysterious AMD crashes. r=sunfish
2016-07-08 17:30:30 +02:00
94e2b5ece5
Bug 1285040 - Adapt device manager pushDir to latest adb; r=bc
2016-07-08 09:20:03 -06:00
54c903fd15
Bug 1285541 - touch clobber to fix bustage
2016-07-08 16:32:49 +02:00
43d6b652f1
Bug 1285541 - Update cubeb to revision b8aebef3. r=padenot
...
Patches by achronop.
MozReview-Commit-ID: 2RurT5Y3CHo
--HG--
rename : media/libcubeb/src/cubeb_audiounit.c => media/libcubeb/src/cubeb_audiounit.cpp
2016-07-08 15:56:56 +02:00
5dba0130cc
Bug 1271751 - part 8 - fix bug in nsTArray_CopyWithConstructors; r=erahm
...
The backwards copying case in MoveOverlappingRegion had a bug: rather
than destroying each element from the source range as we moved it, we
would always destroy the element at the beginning of the source range.
Fortunately, none of the existing types that were copied via
constructors seem to trigger the problematic code.
2016-07-08 16:59:04 -04:00
229ede80e2
Bug 1271751 - part 7 - remove nsTArray_CopyWith*::Copy*; r=erahm
...
After all the previous patches, we never call these functions. Any
copying required by nsTArray is taken care of by other means.
2016-07-08 16:59:05 -04:00
3d1192cadc
Bug 1271751 - part 6 - make ShiftData actually move its elements rather than copying; r=erahm
...
This change eliminates the last use of the Copy* family of functions.
2016-07-08 16:59:05 -04:00
0dc761e84b
Bug 1271751 - part 5 - use MoveNonOverlappingRegion instead of CopyNonOverlappingRegion; r=erahm
...
In all of the calls to CopyNonOverlappingRegion from within nsTArray, we
don't care about the contents of the source afterwards. So we can use
moves instead of copies to potentially make things more efficient.
2016-07-08 16:59:05 -04:00
315612a443
Bug 1271751 - part 4 - rename CopyNonOverlappingRegionWithHeader to MoveNonOverlappingRegionWithHeader; r=erahm
...
Whenever we're copying the header, we can be guaranteed that we're never
going to use the elements from the old array afterward, so can move (in
the C++ sense) the elements rather than copying them.
2016-07-08 16:59:05 -04:00
5dd30fe964
Bug 1271751 - part 3 - add Move{Non,}OverlappingRegion; r=erahm
...
We'll need these for future patches as we transition nsTArray to use
moves for most of its operations rather than copies. The implementation
of these functions are essentially cut-and-paste versions of the Copy*
functions, but using moves.
2016-07-08 16:59:05 -04:00
6e9fa45637
Bug 1271751 - part 2 - rename {Copy,Move}Elements to something more accurate; r=erahm
...
The names {Copy,Move}Elements are based on the use of mem{cpy,move},
respectively. However, I submit that we really want the names to
reflect the C++ operations being done, rather than the underlying
implementation details. So let's rename these to reflect that we are
always copying the elements, and discriminate between the two cases
based on whether the regions being copied overlap or not.
Bring CopyHeaderAndElements along for the ride, as well.
2016-07-08 16:59:05 -04:00
e458be747b
Bug 1271751 - part 1 - provide out-of-class definitions for some nsTArray functions; r=erahm
...
This change enables some of the methods in nsTArray to be lazily
instantiated, particularly the ones that care about whether the element
type is copyable. Since we have a number of places where nsTArray is
used with move-only types, we need to ensure that unless methods
requiring copyability are actually called, those methods are not
instantiated.
2016-07-08 16:59:05 -04:00
09a852e8b4
Bug 1271751 - part 0 - remove heap functions from nsTArray; r=erahm
...
They are unused, except for tests.
2016-07-08 16:59:05 -04:00
Nicolas Silva
Sebastin Santy
Carsten "Tomcat" Book
Timothy Nikkel
Boris Zbarsky
Phil Ringnalda
Ryan VanderMeulen
Mason Chang
Randall Barker
Alexander Surkov
Jim Chen
Jan de Mooij
Geoff Brown
Paul Adenot
Nathan Froyd
