Add 1.2 NN HAL interface for dynamic output shape.
Let notify_1_2() notify output shapes.
Document unspecified dimensions and rank.
Bug: 73506513
Bug: 77234888
Test: NeuralNetworksTest_static
Test: VtsHalNeuralnetworksV1_xTargetTest with 1.2 sample driver
Change-Id: I01108913212d9f4aa47daf2f293ea19259925865
diff --git a/neuralnetworks/1.2/IExecutionCallback.hal b/neuralnetworks/1.2/IExecutionCallback.hal
index 667e0d6..47de1b6 100644
--- a/neuralnetworks/1.2/IExecutionCallback.hal
+++ b/neuralnetworks/1.2/IExecutionCallback.hal
@@ -18,6 +18,7 @@
import @1.0::ErrorStatus;
import @1.0::IExecutionCallback;
+import OutputShape;
/**
* IExecutionCallback must be used to return the error status result from an
@@ -39,10 +40,16 @@
* - DEVICE_UNAVAILABLE if driver is offline or busy
* - GENERAL_FAILURE if the asynchronous task resulted in an
* unspecified error
- * - OUTPUT_INSUFFICIENT_SIZE if provided output buffer is
- * not large enough to store the resultant values
+ * - OUTPUT_INSUFFICIENT_SIZE if at least one output
+ * operand buffer is not large enough to store the
+ * corresponding output
* - INVALID_ARGUMENT if one of the input arguments to
* prepareModel is invalid
+ * @param outputShapes A list of shape information of model output operands.
+ * The index into "outputShapes" corresponds with to index
+ * of the output operand in the Request outputs vector.
+ * outputShapes must be empty unless the status is either
+ * NONE or OUTPUT_INSUFFICIENT_SIZE.
*/
- oneway notify_1_2(ErrorStatus status);
+ oneway notify_1_2(ErrorStatus status, vec<OutputShape> outputShapes);
};