Support RadioError:RF_HARDWARE_ISSUE and RadioError:NO_RF_CALIBRATION_INFO for
setRadioPowerResponse_1_6().
Bug: 170938075
Test: build. VTS on Cuttlefish.
Change-Id: Iaf582c31f439810db34693c61e58ff1f4dfd19fd
The mapper 2.0 gralloc passthrough keeps a set of the native_handle_t
pointers of all imported buffers. This change ensures that the
underlying HAL wrapper freeing the native_handle_t is performed
atomically with respect to accesses to the passthrough's native_handle_t
pointer set.
This fixes a race where a native_handle_t could be reallocated between
being freed by the HAL wrapper and removed from the passthrough's set,
which could then cause the import which happened to reallocate that
native_handle_t to spuriously fail.
Bug: 170798776
Test: Manually very no more spurious import failures in ARCVM's decoder.
Change-Id: I654a44e94adb319f54fb462f4484db414ca1b154
The handle type here is needed in some HALs now (specifically in order
to encode a Surface, which is backed by a handle). This previously
wasn't in Java simply because handles are more common in native code,
and so it wasn't expected to be used.
Fixes: 171087128
Test: build
Change-Id: Ie1eaf50cc8e62932cc553b80c02be434bfbaf3b3
On a handover request failure, the modem can now determine
whether or not to fallback. There is also the option to revert
to the legacy logic. Following the legacy logic is the default
behavior.
Test: FrameworkTelephonyTests
Bug: 161572465
Change-Id: Iad778e83ffc264ee25f57f54ff58532d6a8c5cbf
Merged-In: Iad778e83ffc264ee25f57f54ff58532d6a8c5cbf
This reverts commit 540dc2771a.
Reason for revert: reland because target-level 2 are fixed
Bug: 163419744
Change-Id: I293852d50de9a4eeaeec6ccf4c0290f1c1b0012d
Devices with legacy/O/O-MR1 VINTF won't be able to
update to Android S. Devices launching with pre-O,
O, or O-MR1 will need to update its VINTF to P and
above to work with Android S.
Test: builds
Bug: 163419744
Bug: 171260360
Bug: 171260613
Bug: 171260670
Bug: 171260715
Change-Id: I2a65f5a24c0731df9c4247bf9dd2616c2279042f
We are adding things to them, and in order for core libraries like libui
to use them, the newer versions need to be in the VNDK.
Bug: 170435409
Test: build
Change-Id: I4094240656f357b5ae52194befd0e10627b9fe23
Merged-In: I4094240656f357b5ae52194befd0e10627b9fe23
- add utility function to select non one-shot, non on-change and
non special-report sensors
- apply to NoStaleEvents test case
- remove unnecessary checks related to on-change sensors
b/170315229
Change-Id: I349fe8be35fd1fa6f9ff1e0844b09815006a18b8
Add CAP_BLOCK_SUSPEND to android.hardware.health@1.0 and
android.hardware.health@2.1 service so that when epoll_wait()
is waiting on uevents and timer_fd events it can block suspend
to take the necessary action. Without this permission, eventpoll
is going to ignore the flags EPOLLWAKEUP used in health HAL or
healthloop.
Bug: 169873088
Change-Id: If9b0e1f60e47d7bf54942f589acbfbb883b9bac8
Moved the types to andrdoid.hardware.common.fmq.
This allows modules to depend on the unstable version of the library
without conflicting with modules that depend on the stable version of
android.hardware.common.
Test: atest fmq_unit_tests fmq_test
Bug: 142326204
Change-Id: I3091f52c4e763a15fd9a1b7451a1455a06fb4c30
The fact that these variables are 'FILE' instead of 'FILES' is due to
historical changes, but updating comment to reflect this to avoid
confusion.
Bug: N/A
Test: N/A
Change-Id: I92b685eceaab99c7786384c08076dcf3bd087326
Fix VtsHalWifiSupplicantP2pV1_0TargetTest to skip tests
in such a way that VTS reports the module as complete on
devices that don't support Wifi Direct.
Bug: 170326147
Test: run vts -m VtsHalWifiSupplicantP2pV1_0TargetTest
Change-Id: Ia418c4bb72f4772422d724636453395c4ba9fb2d
This needs to be disabled on Pixel 3, but this is the only place w/
source code visible in AOSP. So, it is implemented as a mechanism.
Bug: 166675194
Test: atest android.hardware.camera2.cts.PerformanceTest
Change-Id: I0d91389432906a917696d27d27b4d6d96e64d09f
This CL creates the following primary sets of functions:
* V1_X::utils::convert(<canonical_type>) -- Converts a canonical type
to the corresponding HAL version type.
* nn::convert(<V1_X_HAL_type>) -- Converts a HAL version type to the
corresponding canonical type.
* neuralnetworks::utils::hasNoPointerData -- Indicates if the object
contains no pointer-based data that could be relocated to shared
memory.
* neuralnetworks::utils::flushDataFromPointerToShared -- Relocate
pointer-based data to shared memory.
* neuralnetworks::utils::unflushDataFromSharedToPointer -- Undoes
`flushDataFromPointerToShared` on a Request object. More
specifically, `unflushDataFromSharedToPointer` copies the output
shared memory data from the transformed Request object back to the
output pointer-based memory in the original Request object.
It also introduces some other minor utility code, including
makeQuantized8PerformanceConsistentWithP, countNumberOfConsumers,
validate, valid, and validatedConvertToCanonical.
Bug: 160667419
Test: mma
Change-Id: I0732e658c1f4ed40cd122f1ca8581fb40b056757
Merged-In: I0732e658c1f4ed40cd122f1ca8581fb40b056757
(cherry picked from commit a685c3dbf4)