wine/gecko
0
0
Fork 0
mirror of https://gitlab.winehq.org/wine/wine-gecko.git synced 2024-09-13 09:24:08 -07:00
Code Issues Packages Projects Releases Wiki Activity
132,758 Commits 1 Branch 165 Tags 1.2 GiB
8f3c6c3bb6
Commit Graph

2063 Commits

Author SHA1 Message Date
Vladimir Vukicevic
8f3c6c3bb6 Backed out changeset fd25fb9e951e and 0f0c080ed1ba (bug 843599) for mac debug assertion failures on a CLOSED TREE 2013-05-27 11:45:59 -04:00
Vladimir Vukicevic
ae0320d5fe b=843599, followup -- use MOZ_WIDGET_GONK, not MOZ_B2G, to fix desktop b2g builds 2013-05-27 11:21:47 -04:00
Vladimir Vukicevic
dfdb68d68a b=843599; use gralloc buffers for WebGL streaming on B2G; r=nical,jgilbert,jrmuizel 2013-05-27 10:12:13 -04:00
Markus Stange
e2c9b54971 Bug 876256 - Restore the previous scissor rect at the end of CompositorOGL::DrawQuad. r=mattwoodrow 2013-05-27 14:48:15 +02:00
Chris Lord
56ccb7b5b7 Bug 870055 - Fix AsynPanZoomController continuing movement when touch point doesn't move. r=kats 2013-05-24 15:53:58 +01:00
Anthony Jones
015375f697 Bug 868047 - Use a PresShell ID to eliminate stale APZC events; r=roc,benwa 2013-05-24 09:43:36 +08:00
Nicolas Silva
f47be48b4f Bug 874726 -split ShmemYCbCrImage into YCbCrImageDataSerializer and YCbCrImageDataDeserializer. r=jrmuizel 
--HG--
rename : gfx/layers/ipc/ShmemYCbCrImage.cpp => gfx/layers/YCbCrImageDataSerializer.cpp
rename : gfx/layers/ipc/ShmemYCbCrImage.h => gfx/layers/YCbCrImageDataSerializer.h
 2013-05-23 09:17:10 +02:00
Markus Stange
67e8c4fec5 Bug 861332 - Ensure that the drawing rect used for DrawWindowOverlay doesn't depend on the window's position on the screen, so that for example the 10.6 window resizer is drawn in the correct place. r=mattwoodrow 
--HG--
extra : rebase_source : d47ea3f871a619bfab065dd0eddb74d8c309fed1
 2013-05-22 11:42:49 +02:00
Matt Woodrow
bb67ab0a14 Bug 874369 - Use normal memory instead of Shmem when only sharing between processes. r=Bas 2013-05-22 15:04:12 +08:00
Matt Woodrow
1134362932 Backout 85d108f0b422 (bug 874369) for causing b2g reftest failures. 2013-05-22 13:18:19 +08:00
Matt Woodrow
ce8fcc565c Bug 874369 - Use normal memory instead of Shmem when only sharing between processes. r=Bas 2013-05-22 11:36:39 +08:00
Matt Woodrow
2c3b027765 Bug 874370 - Make ContentClientIncremental use a no-swap transaction. r=nical 2013-05-22 11:36:38 +08:00
Matt Woodrow
14e553db2b Bug 874367 - Release TextureUpdateRequest memory in the destructor to prevent leaks. r=nical 2013-05-22 11:36:38 +08:00
Ed Morley
a8ea0bbb16 Backed out changeset 1d056bb94292 (bug 779395) for reftest failures on Tegras 2013-05-21 14:03:43 +01:00
David Zbarsky
61bdbeaa88 Fix an incorrect comment, no bug, r=me 2013-05-21 18:22:45 +08:00
David Zbarsky
c0eb3c766a Bug 779395 - Add support for skew transforms to compositor-driven animations r=dbaron 2013-05-21 18:22:44 +08:00
Juan Gomez
62de4f04ce Bug 873486 - Remove wrong C++11 virt-specifier MOZ_OVERRIDE in method definition. r=mattwoodrow 2013-05-20 16:31:07 +02:00
Chris Lord
1b8faf626f Bug 869696 - Use an alternative method to unlock gralloc textures on Adreno (TM) 205. r=bjacob 
Targeting the NULL EGLImage causes slowness on the Geeksphone Peak,
and assumedly, other "Adreno (TM) 205" devices. Achieve the same
effect by deleting the GL texture instead.
 2013-05-15 16:45:08 +01:00
Chris Pearce
44c5698dd1 Bug 872375 - Specify picture region in D3D9SurfaceImage. r=padenot 2013-05-20 13:14:13 +08:00
David Zbarsky
988519c5d2 [Bug 873679] Remove bogus assertion in AsyncCompositionManager.cpp r=roc 2013-05-19 19:17:22 -07:00
Gregory Szorc
e0fa1f0276 Bug 863069 - Part 4: Reorder lists to be alphabetical; r=ted 
--HG--
extra : rebase_source : 87ed9d862bc32824d518438b0be74b47a77c198a
 2013-05-16 15:37:18 -07:00
Ryan VanderMeulen
94deccd7c8 Backed out changeset 94978dab7186 (bug 863069) for build bustage. 
CLOSED TREE
 2013-05-16 13:22:02 -04:00
Gregory Szorc
714eb479c3 Bug 863069 - Part 4: More sorting of existing lists; r=ted 2013-05-16 09:53:27 -07:00
Ted Mielczarek
110919af49 bug 871712 - make MSVC PGO opt-in per-directory, and opt-in in the directories that matter. r=glandium 2013-05-16 09:33:26 -04:00
Robert O'Callahan
7f35fbcd56 Bug 868082. Allow SetIsFixedPosition to be used on layers with transforms. In CompositorParent, use the animated transform value as the base transform in CompositorParent::TransformFixedLayers when it has been set by the animation sampler. r=cwiis 
--HG--
extra : rebase_source : 3b809428bf8740753a98defc30ad7b7d96339442
 2013-05-17 00:34:24 +12:00
Matt Woodrow
1d00cb4c28 Bug 870211 - Add ContentClientIncremental. r=nical 2013-05-16 15:45:43 +12:00
Matt Woodrow
43219a3a35 Bug 870211 - Add new IPC messages to LayersTransaction for doing incremental updates to textures. r=Bas 2013-05-16 15:45:43 +12:00
Matt Woodrow
a1675f0918 Bug 870211 - Add ContentHostIncremental. r=nical 2013-05-16 15:45:43 +12:00
Matt Woodrow
614b6071f2 Bug 870211 - Make TextureHosts support updating from partial surfaces. r=Bas 2013-05-16 15:45:43 +12:00
Matt Woodrow
09f575f155 Bug 870211 - Expose details about possible texture upload limitations in TextureFactoryIdentifier. r=nical 2013-05-16 15:45:43 +12:00
Matt Woodrow
321df54209 Bug 870211 - Tidy up ContentClient interface so that it doesn't depend on ThebesLayerBuffer. r=roc 2013-05-16 15:45:42 +12:00
James Willcox
725e374f66 Bug 871390 - Update WebGL canvases async, not 2D ones r=mattwoodrow 2013-05-14 21:16:07 -04:00
Trevor Saunders
c13459e057 bug 869284 - fix some gcc warnings for windows widgetish stuff r=roc 2013-05-06 22:04:02 -04:00
Matt Woodrow
c6dbea8fc2 Bug 756601 - Use NS_roundf so that it compiles on windows. CLOSED TREE 2013-05-10 09:50:33 +12:00
Matt Woodrow
428c89c1b2 Bug 756601 - Fix bugs with OMTC component alpha. r=roc 2013-05-10 09:02:50 +12:00
Matt Woodrow
1e094458f5 Bug 756601 - Send the values of mScaleToSize/mScaleMode across IPC. r=roc 2013-05-10 09:02:50 +12:00
Matt Woodrow
5ae991f717 Bug 868259 - Fix BasicCompositor rendering. r=roc 2013-05-10 09:02:50 +12:00
Matt Woodrow
de82bbc24f Bug 756601 - Fix a few small layers refactoring bugs. r=Bas 2013-05-10 09:02:49 +12:00
Matt Woodrow
797b08fa8e Bug 756601 - Setup PaintWillResample correctly. r=roc 2013-05-10 09:02:49 +12:00
Matt Woodrow
1983a897d2 Bug 756601 - Use CreateThebesSurfaceAliasForDrawTarget_hack in CopyableCanvasLayer. r=roc 2013-05-10 09:02:49 +12:00
Matt Woodrow
685f8e9501 Bug 756601 - Round texture coords to the nearest texel. r=Bas 2013-05-10 09:02:49 +12:00
Matt Woodrow
088dbe8514 Bug 756601 - Setup our OpenGL surface before compositing. r=mstange 2013-05-10 09:02:49 +12:00
Matt Woodrow
40a4cc142d Bug 756601 - Use the right texture target when allocating texture in TextureHostOGL. r=Bas 2013-05-10 09:02:48 +12:00
Jeff Muizelaar
c75c330cdb Bug 863498. Get the SurfaceTexture transform every draw. r=nical 
ShadowImageLayerOGL::RenderLayer used to call gl()->GetSharedHandleDetails()
directly and uses that result.

This makes us do something similar. As a concequence we also drop the
mTextureTransform member because we are not caching it anymore.

I do wonder if it would make more sense for the content thread to send
the transform along with the update to make sure that it always matches.
This would also make it so that we have no chance of causing a GC
in the compositor to get the transform by calling through java.

--HG--
extra : rebase_source : 7520870d8f0f0bb81c4acbb3b2011f8b0056d53f
 2013-05-08 18:04:59 -04:00
Ryan VanderMeulen
cb01ebf601 Merge m-c to inbound. 2013-05-09 13:21:14 -04:00
Michael Wu
5555ab54e1 Bug 870375 - back out bug 869011 for startup crashes on devices 2013-05-09 13:01:22 -04:00
Michael Wu
6cbfbe29e1 Backed out changeset 7e4d9be02b8f (bug 869011) for causing bug 870375 2013-05-09 13:00:14 -04:00
Ed Morley
d1df1cdf56 Backed out changeset a23a23055e13 (bug 863498) for compilation failures on a CLOSED TREE 2013-05-09 17:31:59 +01:00
Jeff Muizelaar
d2af6fb834 Bug 863498. Get the SurfaceTexture transform every draw. r=nical 
ShadowImageLayerOGL::RenderLayer used to call gl()->GetSharedHandleDetails()
directly and uses that result.

This makes us do something similar. As a concequence we also drop the
mTextureTransform member because we are not caching it anymore.

I do wonder if it would make more sense for the content thread to send
the transform along with the update to make sure that it always matches.
This would also make it so that we have no chance of causing a GC
in the compositor to get the transform by calling through java.
 2013-05-08 18:04:59 -04:00
Jim Mathies
5e27585b2a Bug 830347 - Fixup Windows omtc code so that it builds builds with the 601 sdk. r=bas 2013-05-09 10:32:50 -05:00