These are required to build. Separate CL to preserve blame history.
From running 'update-includes.sh' script here.
Bug: 37280010
Test: boot/build (sanity)
Change-Id: Id9c2dc37f674b27b947a9f823bd38a550d64d31b
This splits headers into three locations:
include - for backwards compatibility, the global include
include_all - for things system/vendor both use
include_vendor - for things that only vendors use
The goal is to gradually have system things stop referencing
(at least most) of these headers.
Bug: 37280010
Test: build (CL on top adds back in symlinks)
Change-Id: Ibf194276b7faa857e1e7605d7719f4e7d873ecba
getClientTargetProperty will give hardware composer the ability to request some
properties of the client target that hardware composer wants. Prior to this
API, the client will does its best to produce the client target of which the
properties are pretty much fixed.
BUG: b/145968912
Test: boot
Change-Id: Ib8ba44d274a0c8a86c3fde1d09b72f592445ee91
Add a callback to notify the client it should retry a
setActiveConfigWithConstraints call in case of SEAMLESS_NOT_POSSIBLE error.
Test: rev up composer to 2.4
Bug: 141329414
Change-Id: I102bedc9cb89f9d1886a431f8804897bb838eb8f
Based on Forrest run with earlier base build 6069142, the build with this submission failed apct/bluetooth/instrumentation_test
Reason for revert: Break tests
Change-Id: Icccbdafc14394349baf6d116d6590552ebcbeba9
- getSupportedDisplayVsyncPeriods was removed as the vsync period
can be obtained by getDisplayAttribute for each individual config.
- renamed setActiveConfigAndVsyncPeriod -> setActiveConfigWithConstraints
to better describe what this function does.
- added VsyncPeriodChangeTimeline and onVsyncPeriodTimingChanged
Test: rev up composer to 2.4 and test refresh rate switching
Bug: 141329414
Change-Id: Ifec51c893bb9038e9ac48d02c1c7b0fe9e25c108
Add new functions to improve vsync period switching by the platform:
- Adding a list of supported vsync periods to hwc2_config_t to avoid the need to expose
separate hwc2_config_t for each vsync period.
- Adding an API to set the vsync period with timeline constraints to allow better
synchronization with vsync period change.
- Extending HWC2_CALLBACK_VSYNC callback to provide the current vsync period.
Test: rev up composer to 2.4 and test refresh rate switching
Bug: 141329414
Change-Id: I0e3371dc41c4fbcf97073a120307f7a8923c9325
Brightness is already a per display capability, we don't need this API.
BUG: 130313275
Test: build
Change-Id: If1ddf2a6224c9aa86f250178fc5cae3b01334b57
Previously, capability like SkipClientColorTransform is a global capability of
device. However, the hardware is not necessarily able to apply for all
displays, for example external display. Thus, in this patch, we introduce
DisplayCapability for per display capability query.
BUG: 69911676
Test: Build, flash and boot
Change-Id: Id34c83acacb94ebfe27077122e4581522aca9d03
Add a graphics.composer@2.3 interface that will expose
color sampling hardware present on some devices to the
framework. Adds:
getDisplayedContentSamplingAttributes
setDisplayedContentSamplingEnabled
getDisplayedContentSample
Test: boot up pixel3
Test: VtsHalGraphicsComposerV2_3TargetTest on revved Pixel3 hwcomposer
Bug: 116028976
Change-Id: Iab92d2d173f9ae8723f87b75c5e47db8f72baff4
Fixes a couple of problems with the return codes in the readback
documentation and adds a paragraph of clarification about when
getReadbackBufferAttributes will be called.
Bug: 67048889
Test: N/A, doc fix only
Change-Id: I3ea1a6993323f7ce12c36a377bf69e2866b18082
This patch adds the struct of float color, and setLayerFloatColor API so that
SurfaceFlinger can set the layer color with more bits on each channel.
Currently the display calibration tool require 10-bits support.
BUG: 69970838
Test: make locally and applied on device
Change-Id: I8bab335cc4fbc41a3eb18306a8831d1e9a9a219a
hardware/libhardware/include was being included with -isystem, which
hides warnings. Use a macro to convert modules to hw_device_t** to call
the open function that works in C++ and C.
Bug: 32018017
Test: m -j
Change-Id: If68bf15581975f1217fcab366cef7bc784894641
hardware/libhardware/include was being included with -isystem, which
hides warnings. -Wswitch requires that all cases in a switch statement
on an enum be enum values. For enums being used as bit flags, cast the
enum to an int.
Bug: 32018017
Test: m -j
Change-Id: Idb430c451d017bb7ec61a7119d6abd87b6dcdc87
Adds HWC2_CAPABILITY_SKIP_CLIENT_COLOR_TRANSFORM, which indicates
that the client should never apply the color transform, even if all
layers are being composed by the client. This allows the device to
use its color management hardware in all cases to avoid any potential
mismatches between client and device application of the transform.
Bug: 29870302
Change-Id: I90234d91858206714e650eadb03227a4a180d975
Redefines the retire fence returned by presentDisplay to be a present
fence with more strictly defined semantics: this fence should fire
at the vsync when the frame is picked up by hardware, either for
scanout (in the case of video-mode panels) or for transfer to panel
memory (in the case of command-mode panels).
Bug: 29771461
Change-Id: Idf5f86c92a8748426cf3368242c08876a48d81dd
Adds a format in/out parameter to createVirtualDisplay so that the
device knows what format the consumer prefers, but can override it if
it is not capable of rendering to it.
Also adds a surface damage region to the setClientTarget call.
Bug: 22767098
Change-Id: I58801aed2d95fd7dbf63db2e303bc3079caaba25
Documentation-only change that clarifies that in the case where the
client isn't compositing any of the layers, the buffer handle
provided to setClientTarget may be null, and that this should not be
treated as an error by the device.
Bug: 28569722
Change-Id: Ia5825eb0a80c7f838894642654db7d7be14a3b6b
