Commit Graph

13291 Commits

Author SHA1 Message Date
Kartikaya Gupta
9fe7f9b5c0 Back out bug 1121871 as it is an improper fix and creates regressions like bug 1132741. r=BenWa (on IRC) 2015-02-17 16:30:53 -05:00
Carsten "Tomcat" Book
e121494bac Merge mozilla-central to mozilla-inbound 2015-02-17 14:47:52 +01:00
Carsten "Tomcat" Book
f31584a15c Backed out changeset 6952d9fe4f36 (bug 1129871) for causing R14/R15 Test failures 2015-02-17 14:36:42 +01:00
Carsten "Tomcat" Book
a3298e7ebb Backed out changeset e1bb91f60697 (bug 1129871) 2015-02-17 14:32:15 +01:00
Carsten "Tomcat" Book
ad716b71e4 Backed out changeset d81339c7ac52 (bug 1128934) 2015-02-17 14:32:11 +01:00
Jim Mathies
c94fc8510c Bug 1132592 - Avoid showing PluginWidgetParent widgets before they are positioned and clipped. r=aklotz 2015-02-13 12:27:56 -06:00
David Anderson
024881b68a FrameMetrics comparator does not include the scroll generation. (bug 1132715, r=kats) 2015-02-15 13:01:00 -08:00
Kartikaya Gupta
1263c2a53c Bug 1133083 - Fix build when APZCCallbackHelper logging is enabled. r=botond 2015-02-16 21:34:42 -05:00
Kartikaya Gupta
0f981d0fa8 Bug 1131840 - Propagate the EventRegionsOverride flag on RefLayers also. r=roc 2015-02-16 21:30:02 -05:00
Kartikaya Gupta
a2d821499e Bug 1131840 - Add an EventRegionsOverride flag to force an empty hit region. r=roc,botond 2015-02-16 21:30:02 -05:00
Kartikaya Gupta
1f5d00564d Bug 1131840 - Replace the shouldForceDispatchToContent flag with a bitset enum. r=roc,botond 2015-02-16 21:30:02 -05:00
James Willcox
7195045365 Bug 1131793 - Disable SurfaceTexture detach on Mali r=jgilbert 2015-02-14 11:23:01 -06:00
Nicolas Silva
d0462d286c Bug 1128934 - Avoid calling GetClientBounds from BasicComposior. r=roc, karlt 2015-02-14 12:37:32 +01:00
Nicolas Silva
0dff182e54 Bug 1129871 - Favor passing tiles by ref rather than copy part 2. r=BenWa 2015-02-14 12:14:59 +01:00
Nicolas Silva
86cda7d880 Bug 1129871 - Favor passing tiles by ref rather than copy part 1. r=BenWa 2015-02-14 12:14:54 +01:00
Jeff Gilbert
ea33309bb2 Bug 1124394 - Support Core profiles for GLContext. - r=kamidphish 2015-02-12 19:00:41 -08:00
Mason Chang
b374a58cfe Bug 1127151. Part 3: Create a vsync thread loop with dwmflush. r=jrmuizel 2015-02-13 17:17:26 +08:00
Mason Chang
7281b963d7 Bug 1127151. Part 2: Create a vsync source on windows. r=kats 2015-02-13 17:17:24 +08:00
JerryShih
1476642368 Bug 1129686 - fix friend class declaration. r=jrmuizel 2015-02-12 22:26:00 +01:00
Nicolas Silva
fd7dd096b2 Bug 1120331 - "crash in mozilla::ipc::MessageChannel::Send(IPC::Message*)". r=sotaro 2015-02-04 07:15:00 +01:00
Mason Chang
ec9d8d1759 Bug 1130681. Part 1: Add unlock back to ClientTiledLayerBuffer for causing emulator R14 intermittent but doesn't seem necessary. r=jrmuizel 2015-02-13 16:36:46 +08:00
Mason Chang
14d3259912 Bug 1130681. Part 2: Backout bug 1118876 for android reftest failures. r=jrmuizel 2015-02-13 16:36:45 +08:00
Matt Woodrow
1f8d1ab89b Bug 1127925 - Lazily open shared handles in DXGITextureHostD3D11 to avoid holding references to textures that might not be used. r=jrmuizel 2015-02-13 14:23:43 +13:00
zhoubcfan
40d6e6f995 Bug 1129335 - Add private destructor to D3D9 texture usage reporters. r=bgirard 2015-02-16 12:37:43 +01:00
Botond Ballo
898f44f6bb Bug 1127066 - Implement ChromeProcessController::HandleLongTapUp(). r=kats
--HG--
extra : source : 5857a51daa23da0090b92467cc7fe5c23354b981
2015-02-09 18:20:15 -05:00
Botond Ballo
a3f5ee9fb5 Bug 1127066 - Use (nsBaseWidget's) APZEventState in ChromeProcessController. r=kats
--HG--
extra : source : 65d5458b6f6af13ebaf424ccef87d4a0654646c4
2015-02-09 19:06:06 -05:00
Botond Ballo
cd91ae9e1b Bug 1127066 - Complain loudly if APZEventState is used with a widget that doesn't support weak references. r=kats
--HG--
extra : source : e4337bd21059a0ecbffa5d0d567dd9a8f0a3bcd1
2015-02-09 14:10:08 -05:00
Botond Ballo
74667af216 Bug 1127066 - Use a weak reference to the widget in APZEventState to avoid reference cycles. r=kats
--HG--
extra : source : fa8926018068e8de68b0b9c44ad0bdc87acf2033
2015-02-09 14:09:20 -05:00
Botond Ballo
3990a194be Bug 1127066 - Extract an APZEventState class from TabChild. r=kats
--HG--
extra : source : fc54993e0c836ca01f3900072227938e8c4b938c
2015-02-09 14:05:18 -05:00
Botond Ballo
2a72a77d84 Bug 1127066 - Extract SendSetTargetAPZCNotification and its helpers into APZCCallbackHelper. r=kats
--HG--
extra : source : 930454419509346f1002ea7486a128072688a921
2015-02-06 18:11:19 -05:00
Botond Ballo
412107100c Bug 1127066 - Extract a helper to apply the APZ callback transform to a touch event. r=kats
--HG--
extra : source : d938af56c1d8225d794f879a5771c9a17d6a402b
2015-02-05 18:02:27 -05:00
Botond Ballo
eaf4eff1d2 Bug 1127066 - Fold APZCCallbackHelper::UpdateCallbackTransform() into ScrollFrame(). r=kats
--HG--
extra : source : 5b78d9e9ba84408c898d9e0ab55ae30f437792c8
2015-01-28 17:40:29 -05:00
Botond Ballo
98957944f7 Bug 1127066 - Factor out code common to APZCCallbackHelper::UpdateRootFrame and UpdateSubFrame. r=kats
--HG--
extra : source : fe1ea639cdfde510a2bb51d2f113fb14809b6b43
2015-01-23 18:54:19 -05:00
Matt Woodrow
665e47aabe Bug 1131808 - Avoid trying to allocate a buffer for 0 sized YCbCr images. r=nical 2015-02-12 11:23:56 +13:00
Nicholas Nethercote
74cc61fdfd Bug 1131901 (part 1) - Make PL_DHashTableAdd() infallible by default, and add a fallible alternative. r=froydnj.
I kept all the existing PL_DHashTableAdd() calls fallible, in order to be
conservative, except for the ones in nsAtomTable.cpp which already were
followed immediately by an abort on failure.

--HG--
extra : rebase_source : 526d96ab65e4d7d71197b90d086d19fbdd79b7b5
2015-02-02 14:48:58 -08:00
Carsten "Tomcat" Book
34b660c646 merge mozilla-inbound to mozilla-central a=merge 2015-02-11 14:58:16 +01:00
Kartikaya Gupta
1d240ac9cc Bug 1125422 - Read the force-dispatch-to-content flag from the layer tree and use it in the APZ code. r=botond 2015-02-10 16:28:07 -05:00
Kartikaya Gupta
566f0c5f51 Bug 1125422 - Add a flag on ContainerLayer to indicate when the entire subtree of layers needs to be in the dispatch-to-content region. r=roc 2015-02-10 16:28:07 -05:00
Joel Maher
d30568d952 Bug 1125509 - [backout 70aa258394bf] 10% tsvgx linux* regression on inbound (Jan 22) from push c6ece3462afd. r=froydnj 2015-02-10 11:13:34 -05:00
Martin Stransky
6727c7a42d Bug 1097897 - Add HiDPI support (scale) for Gtk2 and update and fix calculation of scaled screen borders. r=karlt 2015-02-10 02:14:00 -05:00
Benoit Girard
81bb5d08d8 Bug 1131462 - Remove buggy GetAsSurface override. r=jrmuizel 2015-02-10 23:03:15 +08:00
Kartikaya Gupta
d5c9c04107 Bug 930939 - Add a function to encapsulate running things on the APZ controller thread. r=botond 2015-02-10 08:24:23 -05:00
Kartikaya Gupta
b53fb2a07c Bug 930939 - Move thread assertion code into APZThreadUtils and add more assertions. r=botond 2015-02-10 08:24:23 -05:00
Milan Sreckovic
dd5cbcf0a9 Bug 1130971 - Null pointer checks before dispatching. r=mattwoodrow
--HG--
extra : rebase_source : f38f8831801730d4e9faf5701ad6210717a2fbc4
2015-02-09 17:23:09 +08:00
JerryShih
af362f98b4 Bug 1129686 - Update cms profile in SRGBOverrideObserver callback. r=jmuizelaar
--HG--
extra : rebase_source : 9fc25be48d86379004ed4fda37c8a9eaf81409ed
2015-02-06 00:18:00 +01:00
Jonathan Kew
ee276d8a7b Bug 1127935 - Check for null when trying to load 'head' table, to avoid crashing on non-sfnt fonts. r=jdaggett 2015-02-10 09:18:23 +00:00
Benoit Girard
af82915f3d Bug 1085223 - Comment out occlusion culling for now. r=mattwoodrow 2015-02-10 11:20:36 +08:00
Nicholas Nethercote
6ee3666899 Bug 1127201 (attempt 2, part 1) - Replace most NS_ABORT_IF_FALSE calls with MOZ_ASSERT. r=Waldo.
--HG--
extra : rebase_source : 488e401ff87e31a2074c4108c4df0572d9536667
2015-02-09 14:34:50 -08:00
Markus Stange
f6e80715ad Bug 1092634 - Let feColorMatrix and feComponentTransfer generate output outside their input's bounds. r=Bas
This makes the color matrix and component transfer Moz2D filters generate
an infinite output, which is then cropped to the primitive's filter
primitive subregion by a subsequent crop filter node. This still gives us
different behavior than other browser when the primitive subregion is
overridden using the x/y/width/height attributes - other browsers either
ignore those completely (IE) or only let them crop the default subregion
(which is defined to be the same as the input subregion) and not enlargen
it - but I'll fix that in a separate bug.
2015-02-09 14:04:11 -05:00
Bas Schouten
555ec83c49 Bug 1118322: Block explicitly for vblank when presenting when using WARP. r=jrmuizel 2015-02-09 16:35:48 +01:00