Nicholas Cameron
d5a16985b4
Bug 934250. Ensure we have a buffer-on-white before using it (ThebesLayerBuffer). r=mattwoodrow
2013-11-11 10:30:51 +13:00
Jim Mathies
be2f3bd4e2
Bug 918273 - Add an apzc tree manager helper for translating device to gecko coordinate space, and use it when sending events to the dom in cases where content is consuming touch. r=kats
2013-11-10 10:43:18 -06:00
Jim Mathies
18fbf3543f
Bug 918273 - Pass back the target scrollable layer id from apzc event methods. r=kats
2013-11-10 10:43:17 -06:00
Kartikaya Gupta
6b0f142e83
Bug 902505 - Adjust the hit-test and input transformations in APZCTreeManager to be more correct. r=botond
2013-11-08 13:56:31 -05:00
Kartikaya Gupta
77617e3aff
Bug 935624 - Update various properties on the FrameMetrics on a layers update. r=botond
...
In particular, in the case with nested APZCs, the parent APZC might request
a repaint at a different resolution, and the child APZC might only find out
about it when the call to NotifyLayersUpdated comes in. At this point the
child APZC must pick up the new values of various properties in order to
remain consistent with the state in gecko and the parent APZC
2013-11-08 11:23:47 -05:00
Kartikaya Gupta
1e20a9603e
Bug 935624 - Kill SetZoomAndResolution because the APZC should not touch the mResolution or mCumulativeResolution. r=botond
2013-11-08 11:23:46 -05:00
Jeff Muizelaar
9344e3053c
Bug 929471. Make CGLTexImageIOSurface2D take a CGLContextObj instead of a NSOpenGLContext. r=bgirard
2013-11-07 22:53:54 -05:00
Matt Brubeck
ee462a968a
Bug 927033 - Always decrement mTouchCount on touchend/cancel [r=kats]
2013-11-07 13:12:33 -08:00
Matt Brubeck
57e61daadc
Bug 927033 - Refactor some duplicate code in APZCTreeManager [r=kats]
2013-11-07 13:11:34 -08:00
Matt Brubeck
cac3956706
Bug 927033 - Don't reset mApzcForInputBlock until all touches are gone [r=kats]
2013-11-07 13:10:29 -08:00
Matt Woodrow
af9108eea2
Bug 35380 - Part 5: Remove aOffset from everywhere. r=nrc
2013-11-07 22:53:08 +13:00
Matt Woodrow
c5ad2001e3
Bug 935380 - Part 4: Make the Compositors use the RenderTarget's origin instead of aOffset. r=nrc
2013-11-07 22:53:08 +13:00
Matt Woodrow
8cd4853004
Bug 935380 - Part 3: Make CompositingRenderTarget store it's origin. r=nrc
2013-11-07 22:53:08 +13:00
Matt Woodrow
73c776bcd2
Bug 935380 - Part 2: Pass the actual rect that the RenderTarget represents as the first parameter to CreateRenderTarget. r=nrc
2013-11-07 22:53:07 +13:00
Matt Woodrow
6ed4f9ac18
Bug 935380 - Part 1: Remove INIT_MODE_COPY since it was invalid for callers to use it. r=nrc
2013-11-07 22:53:07 +13:00
Wes Kocher
9d979344d8
Merge b2g-inbound to m-c
2013-11-06 18:20:55 -08:00
Wes Kocher
29edb85b2f
Merge inbound to m-c
2013-11-06 18:10:46 -08:00
Ryan VanderMeulen
511de9eb44
Merge m-c to b2g-inbound.
2013-11-06 15:00:25 -05:00
Ryan VanderMeulen
267f1bd85e
Merge inbound to m-c.
2013-11-06 14:24:58 -05:00
Sotaro Ikeda
186371b1b2
Bug 934106 - Reduce ImageClient holding video frame number to one. r=nical
2013-11-06 12:32:12 -05:00
Chris Lord
4a7fa8225d
Bug 881451 - Remove unused members of TiledLayerProperties. r=BenWa
2013-11-06 13:42:02 +00:00
Matt Brubeck
acbe123ece
Bug 934750 - Don't cancel content touch events unless APZC consumes them [r=jimm,kats]
...
This is necessary for our "cross-slide" hack to work. We need to send
input to both content and APZC until the input passes the pan threshold, so
that the start page can use touch events without disabling async panning.
In the long run we hope to use "touch-action: pan-x" for this and no longer
rely on this hack.
2013-11-06 12:06:46 -08:00
Ryan VanderMeulen
8a107c2154
Merge m-c to inbound.
2013-11-06 14:59:07 -05:00
David Anderson
91894cb959
Don't copy surfaces in DeprecatedTextureHostBasic::Lock (bug 935641, r=mattwoodrow).
...
--HG--
extra : rebase_source : e97318912cc295b6b1dc5027eba1690e42d24003
2013-11-06 11:55:58 -08:00
Wes Kocher
d3d614d652
Merge m-c to inbound
2013-11-04 21:22:23 -08:00
Wes Kocher
32a09c5f75
Merge inbound to m-c
2013-11-04 21:09:17 -08:00
Matt Woodrow
56073d6b1a
Bug 923434 - Use DrawSurface in ThebesLayerBuffer instead of FillRect. r=Bas
2013-11-05 17:50:57 +13:00
Matt Woodrow
d983ecfd5c
Bug 923434 - Don't use operator source for D2D in ThebesLayerBuffer since it's slower. r=Bas
2013-11-05 17:50:56 +13:00
Wes Kocher
f0c23390fe
Backed out changeset bad6c063b01f (bug 932274) for build bustage
2013-11-04 14:40:50 -08:00
Benoit Girard
96dd2833db
Bug 930476 - Show paint markers for async frames. r=ehsan
2013-11-04 17:04:51 -05:00
David Zbarsky
906bff4c2b
Bug 932274: Use Moz2D size classes in ipdl r=nical
2013-11-04 16:43:10 -05:00
Kartikaya Gupta
0114edf4cd
Bug 934439 - Add an APZC_LOG line for UpdateScrollOffset. r=botond
2013-11-04 21:30:10 -05:00
Ryan VanderMeulen
742c6ee1dd
Merge m-c to b2g-inbound.
2013-11-04 16:32:13 -05:00
Botond Ballo
e23a67dc35
Bug 932525 - Better naming for some transformation matrix variables. r=kats
2013-10-31 14:44:33 -04:00
Botond Ballo
fa777434ed
Bug 932525 - Do APZC hit testing using the layer's screen coordinates rather than layer coordinates. r=kats
2013-10-30 15:58:25 -04:00
Nathan Froyd
0796aeb3b7
Bug 918651 - part 9 - remove useless includes from IPDL files; r=ehsan
...
These includes are no longer needed. It's much easier to locate these
includes when they're not obscured by a bunch of other stuff.
2013-10-18 14:47:04 -04:00
Nathan Froyd
ccdd4fb35d
Bug 918651 - part 8 - remove redundant includes from IPDL files; r=ehsan
...
These include statements have been made redundant by all the qualifications
to the using statements.
2013-10-18 13:34:08 -04:00
Nathan Froyd
3a9ed362e7
Bug 918651 - part 7b - fix Layers{Surfaces,Messages}.ipdlh header bootlegging; rs=ehsan
...
LayersSurfaces.ipdlh declares itself as |using| a bunch of types that only
get used in LayersMessages.ipdlh. Since intelligent include-placing
recognizes that these types are not used in LayersSurfaces.h, the necessary
header lines get left out of the generated LayersSurfaces.h, leading to
compilation problems on some platforms. Fix this by moving the appropriate
lines to LayersMessages.ipdlh.
2013-10-19 12:09:59 -04:00
Nathan Froyd
e2ede970d2
Bug 918651 - part 5 - qualify all IPDL using statements with required header; r=ehsan
2013-10-01 15:25:07 -04:00
Nathan Froyd
973f5e7322
Bug 918651 - part 1 - remove unnecessary using statements from IPDL files; r=ehsan
2013-10-18 14:48:13 -04:00
Mike Hommey
91865c8cd6
Bug 932618 - Use NO_PROFILE_GUIDED_OPTIMIZE to disable PGO for individual files. r=gps
2013-11-02 08:42:53 +09:00
Ed Morley
aa914d52d6
Backed out changeset f24ee0606ea6 (bug 926128) for reftest failures on OS X & crashes on B2G on a CLOSED TREE
2013-11-01 15:51:32 +00:00
Andreas Gal
4309023622
Bug 926128 - Use a single configurable shader in OpenGL layers backend. r=vlad, r=nical
2013-11-01 09:29:20 -04:00
Ed Morley
d1eeb9782e
Merge mozilla-central and inbound
2013-11-01 13:18:19 +00:00
Ed Morley
641a5b27d4
Merge latest green b2g-inbound changeset and mozilla-central
2013-11-01 13:14:45 +00:00
Benoit Jacob
46729cad0c
Bug 886219 - Avoid introducing a global constructor with the ImageBridgeParent singleton - r=nical
2013-11-01 07:34:40 -04:00
Fabrice Desré
e5095467d5
Bug 914843 - fix gfx issue on oop b2g linux. r=nrc
2013-10-31 23:36:02 -07:00
Ryan VanderMeulen
c80a525c6d
Merge m-c to inbound on a CLOSED TREE.
2013-10-30 22:42:13 -04:00
Kartikaya Gupta
78cc3658d4
Bug 906877 - Remove code obsoleted by multi-apzc. r=BenWa, vingtetun
...
Now that we have multi-APZC landed, we don't need BrowserElementPanning
to detect when it's about to scroll a subframe and let APZC know. So
the two messages involved in this can be completely removed.
2013-10-30 17:06:15 -04:00
Kartikaya Gupta
61ffd37c9d
Bug 732971 - For Fennec move the resolution from the top-level presShell to the content presShell. r=BenWa
...
The bulk of this patch is fixing up pieces of code that infer the resolution
as the inverse of the scaling transform on the root layer. This used to be
how various bits of gfx code obtained the resolution, but now that we can set
the resolution on the actual presShell that contains the content, we can also
just read the resolution from the FrameMetrics directly.
2013-10-29 09:27:50 -04:00
