Also some updates to HIDL interface:
- Add releaseRecordingFrameHandle to ICameraDevice
for native handle metadata recording mode
- Add handleCallbackTimestamp to ICameraDevieCallback
for native handle metadata recording mode
- Add missing face detection metadata to
ICameraDeviceCallback::dataCallback
- Instead of passing native handle, pass buffer ID
in dequeueBuffer/enqueueBuffer/cancelBuffer in
ICameraDevicePreviewCallback
- Add CameraFrameMetadata in types.hal for face
metadata
Test: Camera CTS passing (except FlashLightTest) on Angler
Bug: 30985004
Change-Id: Idf72a4b5f4c934845ac698f0b13536608ffd0100
This fixes a perf/feature regression from the legacy HAL definitions;
previously, vendor implementations could use custom values with this
field to implement custom features on their devices.
Add several enum values that can be used for this purpose.
Test: New CTS test passes
Bug: 34853980
Change-Id: I1b37a9511326178b8eb7536ff99e14d83be3b0c2
This SELinux context doesn't have permission to access stdout
in read/write mode. Doing so will result in violation and
subsequent failure when running in binderized mode. Switch to
'/dev/null' which still can be used with SELinux enabled.
BUG: 35701060
Test: adb shell
/data/nativetest/VtsHalCameraProviderV2_4TargetTest/VtsHalCameraProviderV2_4TargetTest
Change-Id: Ib1cf2766ca41cfe3600ad8a4442a334dfab0e2c1
Use single camera capture requests to verify basic
'processCaptureRequest' functionality:
- 'processCaptureRequestSinglePreview' will generate
a valid preview capture request. The result needs to
include both valid stream Id and frame number.
- 'processCaptureRequestInvalidSinglePreview' will
omit the settings from the first capture request. Hal
should handle this by returning an appropriate error.
- 'processCaptureRequestInvalidSingleSnapshot' will
have a valid blob request but no valid output buffers.
Hal should again return appropriate error in this case.
BUG: 32022758
Test: compile and run the gtest binary on device
Change-Id: I021dd150b12d4be39fae47e13ba82d3db105bfa3
Further tests concerning camera configurations. The
additional tests corver:
- The available stream configurations.
- Invalid stream parameters.
- Constrained mode if available.
- ZSL mode if available.
- Stream combinations including preview and still capture.
- Stream combinations including video and snapshot.
BUG: 32022758
Test: compile and run the gtest binary on device
Change-Id: I5111ac96b4aaa7ad9f163f990f6b0d0c229993f9
Native buffer handles should always get closed
and deleted during buffer release.
BUG: 35338606
Test: Manual using application
Change-Id: I5d7a4b02a28b29b872b66843d623166f04207c2e
We need google shims on the vendor partition because they are providing
an implementation of a vendor defined interface. They were written by
google just as a courtesy/to make the transition easier. They're
basically a set for vendors to assemble their hal implementations
from.
Bug: 34135607
Test: marlin persist.hal.binderization on/off
Change-Id: I2e2af5af39264cf290259755bb9b2eb9827a21f5
And definitely don't stop init if there are no vendor tags defined at
all.
Instead, just log about it.
Test: gce camera HAL init succeeds; no regression to sailfish camera
Bug: 35036220
Change-Id: I7f4481d7ec1a88df0c67a66c364f351b3321b221
If the torch callback is fired during CameraProvider
constructor, the sp will cause it being deleted
when the callback returns.
Test: fix the crash
Bug: 35216523
Change-Id: I133d866ed603d6c7e549b57d20a911af7a7e0698
Java intermediates are now created in the common rather than the device-
specific generated sources subdirectory.
Bug: 33842609
Test: make
Change-Id: I16af4cd19bf0b24baa0ad9ebfdb86b4d959c8672
The stream resources will be released by Hal most
likely only in cases of successful configuration.
If the configuration fails keep the deleted streams
for the next call.
BUG: 34131351
Test: 'CameraTest' API1 tests using Hal3.x and ZSL
Change-Id: I68696d561258571727b35b52ff326aac27edaad3
No longer needed. This also enable us to switch to
defaultPassthroughImplementation.
Also rename camera device wrapper library to signal it's not a
top level HAL entry.
Test: marlin camera app with passthrough/binderzied mode
Bug: 34817742
Change-Id: I78ad5afbf78b2679be3efe0e82f430958ef6bfea
Camera is currently compiled separately in passthrough and binderized
modes. Until that is resolved, adding a call to internal hidl logic.
Bug: 34817742
Test: camera starts up and works and respects binderization toggle in
both modes on marlin.
Change-Id: I20733cbc82a68a1da1c4f1b3754aef2887ca718a
This is a new public API feature for O, which needs to be plumbed
down to the HAL layer as well.
Test: New CTS tests pass
Bug: 30983997
Change-Id: Ie9f8a37cb0b149f7114f11fcf611a43cac60abe7
This will introduce one more FD dup/close in passthrough mode, but
we have no other choice as camera server will not know whether
it is running in passthrough or binderized mode.
Test: running Camera2 API CTS
Bug: 30985004
Change-Id: I407d54a51c51fc09477dfb6236ab1b53bb1e0a21
This gives each buffer a unique identifier and allow camera server
to only send one unique to HAL process once.
Bug: 30985004
Change-Id: I6aff498c5667dd71fd70b3881fd60b0ecc366e0a
Right now conventional HAL close the acquire_fence FD so wrapper
should not close it again. However, we might need to change this
behavior later to make camera service agnostic to passthrough or
binderized mode.
Bug: 34169301
Change-Id: I96b46521e4081b43e7a7e9c9efa98121739dda99
Bug fixes like deadlock resolution, wrong enum usage etc.
Bug: 30985004
Test: run Camera2 API CTS tests on Angler
Change-Id: I661fa9197f66344ddecca8f68d343c891806eca1
Fully covered: ICameraProvider@2.4 ICameraDevice@3.2
ICameraDeviceSession@3.2 is partially covered.
(configure_stream/process_capture_request tests to be added later)
Test: compile and run the gtest binary on device
Bug: 30985004
Change-Id: I61f20d14a11bd69fc19b06062c489dd8833b09a4
Supports legacy camera HAL modules; also exports ICameraDevice
instances.
Test: compile
Bug: 30985004
Change-Id: I2b9624a412de95dd43979a5e6650b170010c577a
