Commit Graph

237010 Commits

Author SHA1 Message Date
Jean-Yves Avenard
d7b7286d25 Bug 1149343 - Part2: Add HW decode blacklisting support on mac. r=mattwoodrow 2015-03-31 14:18:38 +11:00
Matt Woodrow
b9044f9109 Bug 1149343 - Part 1: Unify prefs/blacklist for hardware acceleration video decoding. r=cpearce 2015-04-01 16:06:56 +13:00
Byron Campen [:bwc]
2f71229f66 Bug 1149298 - Part 3: Fire end of candidates signal when StartGathering has nothing to do, and only call StartGathering once per offer/answer. r=drno 2015-03-30 16:32:51 -07:00
Byron Campen [:bwc]
3773d15550 Bug 1149298 - Part 2: When destroying a candidate, ensure that the ice_ctx doesn't continue waiting for it to init. r=drno 2015-03-30 15:22:28 -07:00
Byron Campen [:bwc]
30912960d8 Bug 1149298 - Part 1: Test cases. r=drno 2015-03-30 15:46:49 -07:00
L. David Baron
900854f647 Backed out changeset e942133b1f9b (bug 1149769) for jittest-2, jsreftest, and spidermonkey test failures, on a CLOSED TREE. 2015-03-31 19:37:54 -07:00
Eric Faust
ef2ada318b Bug 1149769 - Split reflect-parse.js reftest into its own subdirectory. (r=Waldo) 2015-03-31 18:29:10 -07:00
Ben Turner
d6dbe650bd Bustage fix for bug 1131776, s/moz_malloc/malloc/ on CLOSED TREE 2015-03-31 16:27:36 -07:00
Jonathan Watt
785f13213f Bug 1145246, part 7 - Add more CSS animation tests to check currentTime clamping. r=birtles 2015-03-23 13:23:19 +00:00
Jonathan Watt
a985df248c Bug 1145246, part 6 - Remove disabled code from AnimationPlayer::ComposeStyle. r=birtles 2015-03-23 13:23:10 +00:00
Cameron McCormack
6fbf636225 Bug 1145912 - Add some logging to help determine why this test intermittently fails; no review. 2015-04-01 10:08:35 +11:00
Ben Turner
62f61c5626 Bug 1131776 - Use WITHOUT ROWID tables for IndexedDB, r=janv. 2015-02-16 09:48:14 -08:00
Ben Turner
331d4866a7 Bug 1112702 - Make IndexedDB transactions non-durable by default, r=khuey. 2015-01-25 22:30:09 -08:00
Ben Turner
c5c45c8f37 Bug 1131766 - Use a simpler thread model for IndexedDB transactions, r=khuey. 2014-12-02 12:01:44 -08:00
Ben Turner
975ba73ad9 Bug 866846 - Use WAL journal mode for IndexedDB databases, r=janv. 2015-01-24 08:16:26 -08:00
Ben Turner
6424c1255f Revert unintentional change to crash reporting infra in changeset ce2692d64bcf DONTBUILD 2015-03-31 15:51:24 -07:00
Wes Kocher
f33e532cc7 Backed out changeset 6b886fbb3e48 (bug 1148032) for mochitest-chrome orange on b2g emulators 2015-03-31 15:48:25 -07:00
Ben Turner
d839795997 Bug 1112620 - Fix invalidated version change transactions, r=khuey. 2015-03-31 15:35:12 -07:00
Ben Turner
4e377455b4 Bug 1071360 - Fix async storage connection closing when open fails, r=asuth. 2015-03-27 22:45:28 -07:00
Jan Varga
2cf063d71d Bug 1144806, Fix test_temporary_storage to not depend on database sizes. r=bent. 2015-03-25 09:49:37 -07:00
Andrea Marchesini
d39381422e Bug 1143526 - DataStore API should create DOMError with proper error messages, r=fabrice 2015-03-31 23:23:52 +01:00
Jim Chen
e87cbf11c3 Bug 1149189 - Add test for setting the same composing text; r=esawin 2015-03-31 18:20:27 -04:00
Jim Chen
a76efd387b Bug 1149189 - Add dummy text change when setting composition to the same text; r=esawin 2015-03-31 18:20:27 -04:00
Jim Chen
59365aafeb Bug 1149189 - Backout e75d824c730d "Force Gecko update on endBatchEdit for range updates"; r=esawin 2015-03-31 18:20:27 -04:00
Jim Chen
84a2992f8e Bug 1149172 - Query IMEStateManager for composition state; r=esawin 2015-03-31 18:20:27 -04:00
Jim Chen
addd99f09e Bug 1149172 - Send well-formed IME events; r=esawin 2015-03-31 18:20:27 -04:00
Jim Chen
4d4646e939 Bug 1149172 - Fix small bugs in IME focus handshake; r=esawin 2015-03-31 18:20:27 -04:00
Mark Goodwin
bead98d47f Bug 1138848 - Tests for modified OneCRL (r=keeler, unfocused)
* * *
* * *
give blocklist debug info to NSPR_LOG
2015-03-31 15:10:19 -07:00
Mark Goodwin
695994d59d Bug 1138848 - Modify OneCRL blocklist for subject / public key blocking (r=keeler, unfocused) 2015-03-31 15:10:09 -07:00
Mats Palmgren
b9fae4bb36 Bug 1149785 - Make nsIWidget::ClientToWindowSize use LayoutDeviceIntSize instead of the unit-less nsIntSize. r=roc 2015-03-31 22:09:03 +00:00
Mats Palmgren
a0a4fe694d Bug 1149784 - Make nsSubDocumentFrame::GetMarginAttributes use CSSIntSize instead of the unit-less nsIntSize. r=roc 2015-03-31 22:09:02 +00:00
Andreas Tolfsen
25c01263d4 Bug 1149625: Order errors by name rather that status code
This patch has no functional changes, apart from correctly naming
InvalidElementStateError and exporting it.  It's not currently in use.

r=dburns
2015-03-31 18:00:32 +01:00
L. David Baron
42236287fa Bug 847287 patch 13 - Apply animations/transitions on the layer in the order that reflects how they override, rather than the opposite. r=birtles
This independently would have fixed some of the problems fixed in this
bug.  It would not have fixed them fully, though, since it would not
have prevented us from sending animations being covered up by !important
rules to the compositor.
2015-03-31 15:05:55 -07:00
L. David Baron
58f800f801 Bug 847287 patch 12 - Check mWinsInCascade for all callers of GetAnimationOfProperty/HasAnimationOfProperty. r=birtles
This patch (after stepping through the call graph) affects the following
places:
 * CommonAnimationManager::GetAnimationsForCompositor, which is used
   only by nsDisplayListBuilder::AddAnimationsAndTransitionsToLayer,
   which already checks the individual animations (so really no change)
 * AnimationPlayerCollection::CanThrottleAnimation
 * ActiveLayerTracker::IsStyleAnimated
 * nsLayoutUtils::HasAnimationsForCompositor
 * nsLayoutUtils::HasAnimations (which is used only to check whether we
   can make the 0-opacity optimization)
I believe it makes sense to change all of these locations (although in
the long term we want to throttle (or similar) more animations).

Without this patch, I believe we're forcing the creation of an opacity
layer because we think we have animations to send to it.
2015-03-31 15:05:55 -07:00
L. David Baron
4978814ccb Bug 847287 patch 11 - Only check that we can throttle animations once per refresh cycle (or invalidation of style rule), to make it cheaper to call EnsureStyleRuleFor more than once per refresh cycle. r=birtles
This saves some extra work that we don't need to do.

Mechanically, the patch moves a chunk of code that is around the last
part of the function and converts it to an early return that's slightly
earlier than that last part, thus also including the skipping of the
throttling checks in what we skip for the early return.

I want to do this here since patch 9 introduces a new call to
EnsureStyleRuleFor.
2015-03-31 15:05:55 -07:00
L. David Baron
9e47fdcc7b Bug 847287 patch 10 - Only call CheckNeedsRefresh if we might have changed mNeedsRefreshes, to make it cheaper to call EnsureStyleRuleFor more than once per refresh cycle. r=birtles
This saves some extra work that we don't need to do.

I want to do this here since patch 9 introduces a new call to
EnsureStyleRuleFor.
2015-03-31 15:05:55 -07:00
L. David Baron
957e72b3a6 Bug 847287 patch 9 - Use fully-updated style rule for animations when updating cascade results for transitions. r=birtles
I don't have a test case that requires this, but it seems like a good
idea.  (It was an incorrect theory for fixing a test failure that I was
debugging, but still seems worth doing.)
2015-03-31 15:05:55 -07:00
L. David Baron
bb63a8c97c Bug 847287 patch 8 - Only update transition manager's cascade results when an animation starts or stops being in effect. r=birtles
This avoids some extra work that was added in bug 1125455 now that we
have a mechanism for detecting when animations start and stop being in
effect (introduced in patch 7).

This is also needed to prevent infinite recursion in patch 9.
2015-03-31 15:05:54 -07:00
L. David Baron
4989c2adce Bug 847287 patch 7 - Dynamically update cascade results when animations start or stop being in effect. r=birtles
This is an additional part of the main work in this bug; it keeps
mWinsInCascade updated in cases where we need to update it.
2015-03-31 15:05:54 -07:00
L. David Baron
e30b5e2919 Bug 847287 patch 6 - Set mWinsInCascade for CSS Animations. r=birtles
This is the main patch for the bug; it makes us use the mechanism added
in bug 1125455 to avoid sending animations that aren't currently
applying to the compositor.

Patch 7 is needed to make this code rerun in all the cases where we need
to rerun it, though.
2015-03-31 15:05:54 -07:00
L. David Baron
9fd12c96a8 Bug 847287 patch 5 - Add method to update animations on layer. r=birtles
This does somewhat less work than PostRestyleForAnimation, although I
believe PostRestyleForAnimation would be a sufficient alternative.

This is used in patch 6.
2015-03-31 15:05:54 -07:00
L. David Baron
c490cf5044 Bug 847287 patch 4 - Add a method to nsRuleNode that reports the properties overriding a CSS animation. r=birtles
This is used in patch 6.
2015-03-31 15:05:54 -07:00
L. David Baron
222514f971 Bug 847287 patch 3 - Add assertions about consistency of the flags for animating on the compositor. r=birtles
These flags were added in patch 2.
2015-03-31 15:05:54 -07:00
L. David Baron
35c6d8f0dd Bug 847287 patch 2 - Add flag for CSS properties that can be animated on the compositor. r=birtles
Patch 3 adds sanity-checks to check these flags against other equivalent
data.

This flag is used in patch 5 and patch 6.
2015-03-31 15:05:54 -07:00
L. David Baron
f2c8d84ec1 Bug 847287 patch 1 - Add additional tests. r=birtles
All of the todos will be fixed by later patches in this bug (as will
some already-existing todos in the same file).
2015-03-31 15:05:54 -07:00
Zachary Carter
b318ec6dfb Bug 1148565 - Add the default profile image to Aero theme r=markh 2015-03-31 15:03:11 -07:00
Zachary Carter
c7eb99baf5 Bug 1140627: Make FxAccountsCommon.js variable OAUTH_SERVER_ERRNO_OFFSET non-const, to fix JS warning. r=dholbert 2015-03-31 15:01:31 -07:00
Byron Campen [:bwc]
a0d2c7a604 Bug 1148649: Reenable video multistream mochitests on debug e10s. r=drno 2015-03-26 13:03:29 -07:00
Tom Schuster
b6e24288b0 Bug 1147005 - Change JSAddPropertyOp signature. r=jorendorff,peterv 2015-03-28 14:47:02 +01:00
Andrea Marchesini
5c5be3ffed Bug 1148032 - BroadcastChannel API should not bypass private browsing mode, r=ehsan 2015-03-31 22:11:47 +01:00