1. Add setDisplayBrightness as a display command, so that it may be
set atomically with composition updates for that display.
2. Adjust tests to set display brightness using display commands.
The setDisplayBrightness api on IComposer should be deprecated. It will
be removed in a follow-up patch.
Bug: 210151839
Test: builds
Change-Id: I2e4348e1d7f799d1744390afbb9bd206054eb933
The first time this topic landed, it resulted in b/212402133. We avoid
running into this bug with Ib11d46439db57b90486bad07dd90f2cf0822182a.
Remove references to 8 bit alpha.
Original commit message:
After some discussion, we've decided to always set the ScreenDecorations
to DISPLAY_DECORATION. HWC can decide, based on the format, whether to
truly treat it as DISPLAY_DECORATION, or to change to DEVICE
composition.
Update the documentation to reflect this change.
See also I1da3199bacf5877e88876249c60c0a33fcd44c1e.
Bug: 193170859
Test: NA
Change-Id: Iaf5385c38f6827e07b91a9a0a8769fb7a125e847
The test process was crashing on null pointer instead of exiting
successfully, so fixing the null pointer error.
Test: VTS
Change-Id: I1d6988d25ed58c159adc49d43d36e4b3f0ffa289
- No one is using the translate functions, so removing.
- Rename the static_assert portion to Hidl2AidlAsserts
Test: build
Change-Id: I42abc37953e1d927a3bcf5f60f56acf99fde6bf5
HIDL expects that Presented will be 1 and Validated witll be 0.
Even though we map the enum values of PresentOrValidate.Result
to the correct value, this CL makes the interface consistent with
the HIDL so that we can directly use the enum value.
Bug: 208856704
Test: VTS
Change-Id: I9fb2421e32c2288c5495422eda6764f6102337f0
Add @nullable to the releaseFence in IComposerClient.setReadbackBuffer
as it can be null, if the buffer is ready to be consumed.
Bug: 210166943
Test: VTS
Change-Id: Ic556ceabd22ba8acddf37ca7d9a6d14d41697c5b
There are no current users for GenericMetadata, so removing
from composer 3.
Bug: 209691612
Test: VTS
Change-Id: I729a3646d9684c78454708fd15f1571aa405e9bb
Remove the hint from DisplayCommand.colorTransform as currently
we only use ColorTransform::IDENTITY and ColorTransform::ARBITRARY_MATRIX,
which both can be expressed using a matrix.
Bug: 208879711
Test: VTS
Change-Id: Ibb24593e3d0af5afdc8f3f79de2fc22eeccfcea8
After some discussion, we've decided to always set the ScreenDecorations
to DISPLAY_DECORATION. HWC can decide, based on the format, whether to
truly treat it as DISPLAY_DECORATION, or to change to DEVICE
composition.
Update the documentation to reflect this change.
See also I1da3199bacf5877e88876249c60c0a33fcd44c1e.
Bug: 193170859
Test: NA
Change-Id: I1fae74b360d1b9d79fc4afeab6f71089d4c8b7e5
run VtsHalGraphicsComposerV2_2TargetTest will timeout failed
when there is dual displays. use `std::unordered_set` Display type,
`std::vector<Display>(mDisplays.begin(), mDisplays.end())` as
getDisplays return value, mDisplays elements order may not be
same as the order of hotplug events during boot matters.
We should store `mDisplays` as `std::vector`, since the order of
hotplug events during boot matters in determining the primary display.
Bug: 209409863
Test: run vts -m VtsHalGraphicsComposerV2_2TargetTest
Signed-off-by: Kui1 Wu <wukui1@xiaomi.com>
Change-Id: I628ad593640f28fb8a916822a91ac2c578cc299d
These classes are used by both the client and the service, which
makes them confusing. This CL splits the logic for the client
and the service.
Bug: 208856704
Test: VTS
Change-Id: I6fa89858afeee9113ea8c810261d734163a95ec9
Also remove the redundant APIs getDisplayBrightnessSupport
and getDozeSupport.
Bug: 209889459
Test: VTS
Change-Id: Id7e4da62fdcf65b279656a9f115e60b6145e8b5d
This CL also includes a few miscellaneous changes:
- Remove ..composer3.command namespace
(so all commands are in composer3)
- Minor documentation update
- Minor name changes
Bug: 201008710
Test: VTS
Change-Id: I3fc24392d6c019c39e4d7df735abacc2c793c4f9
