Add Quality of Service to NNAPI HAL
This CL makes the following changes:
* introduces a new Priority enum
* extends ErrorStatus with new error codes
* adds "supportsDeadline" method to IDevice
* adds priority and deadline arguments to IDevice::prepareModel*
* adds deadline argument to IPreparedModel::execute*
* updates IExecutionCallback with new ErrorStatus
* updates current.txt accordingly
Bug: 136739795
Bug: 142902514
Bug: 145300530
Test: mma
Change-Id: Iaa7877bde1f463635b8bbdb4e8a001d7b79b9c65
diff --git a/neuralnetworks/1.3/IPreparedModelCallback.hal b/neuralnetworks/1.3/IPreparedModelCallback.hal
index ff295a2..11ebbf4 100644
--- a/neuralnetworks/1.3/IPreparedModelCallback.hal
+++ b/neuralnetworks/1.3/IPreparedModelCallback.hal
@@ -16,7 +16,6 @@
package android.hardware.neuralnetworks@1.3;
-import @1.0::ErrorStatus;
import @1.2::IPreparedModelCallback;
import IPreparedModel;
@@ -48,6 +47,10 @@
* unspecified error
* - INVALID_ARGUMENT if one of the input arguments to
* prepareModel is invalid
+ * - MISSED_DEADLINE_* if the deadline for executing a model
+ * cannot be met
+ * - RESOURCE_EXHAUSTED_* if the task was aborted by the
+ * driver
* @param preparedModel A model that has been asynchronously prepared for
* execution. If the model was unable to be prepared
* due to an error, nullptr must be passed in place of