Documentation improvements.
Specify that an output cannot overlap with an output/input/constant.
Document required lifetime of ANeuralNetworksOperandType* parameters.
Document ANEURALNETWORKS_OEM_OPERATION shape constraints:
Because the semantics of an ANEURALNETWORKS_OEM_OPERATION are
determined entirely by the driver, the NNAPI runtime cannot deduce
the shape of its output(s); and we have not designed the HAL in a way
that allows the driver to to communicate output shape back to the
runtime.
The NNAPI partitioner cannot handle an inter-partition temporary of
unknown shape; and because ANEURALNETWORKS_OEM_OPERATION cannot be
executed on the CPU, the partitioner has no guaranteed way to find a
partitioning where every inter-partition temporary has known shape.
We avoid this problem by requiring that any model containing an
ANEURALNETWORKS_OEM_OPERATION either contain only one operation (so
there is only one partition) or contain only operands of known
shape.
Fix typos.
Test: mma at tree root
Bug: 79888238
Bug: 79152812
Bug: 77236591
Change-Id: I1916f85c79f0942e8c313c4b1f67aea701fe58ba
diff --git a/nn/runtime/include/NeuralNetworks.h b/nn/runtime/include/NeuralNetworks.h
index 834c8c6..498a463 100644
--- a/nn/runtime/include/NeuralNetworks.h
+++ b/nn/runtime/include/NeuralNetworks.h
@@ -1879,10 +1879,10 @@
* <p>To use:<ul>
* <li>Create a new execution instance by calling the
* {@link ANeuralNetworksExecution_create} function.</li>
- * <li>Associate data to the model inputs with
+ * <li>Associate input buffers or memory regions to the model inputs with
* {@link ANeuralNetworksExecution_setInput} or
* {@link ANeuralNetworksExecution_setInputFromMemory}.</li>
- * <li>Associate output buffers to the model outputs with
+ * <li>Associate output buffers or memory regions to the model outputs with
* {@link ANeuralNetworksExecution_setOutput} or
* {@link ANeuralNetworksExecution_setOutputFromMemory}.</li>
* <li>Apply the model with {@link ANeuralNetworksExecution_startCompute}.</li>
@@ -1891,6 +1891,11 @@
* <li>Destroy the execution with
* {@link ANeuralNetworksExecution_free}.</li></ul></p>
*
+ * <p>An output buffer or memory region must not overlap with any
+ * other output buffer or memory region, with an input buffer or
+ * memory region, or with an operand value in a memory object
+ * ({@link ANeuralNetworksModel_setOperandValueFromMemory}).</p>
+ *
* <p>An execution cannot be modified once {@link ANeuralNetworksExecution_startCompute}
* has been called on it.</p>
*
@@ -1903,7 +1908,7 @@
* thread to use {@link ANeuralNetworksEvent_wait} at the same time.</p>
*
* <p>It is also the application's responsibility to ensure that there are no other
- * uses of the request after calling {@link ANeuralNetworksExecution_free}.</p>
+ * uses of the execution after calling {@link ANeuralNetworksExecution_free}.</p>
*/
typedef struct ANeuralNetworksExecution ANeuralNetworksExecution;
@@ -2086,7 +2091,9 @@
*
* @param model The model to be modified.
* @param type The {@link ANeuralNetworksOperandType} that describes the shape
- * of the operand.
+ * of the operand. Neither the {@link ANeuralNetworksOperandType}
+ * nor the dimensions it points to need to outlive the call to
+ * {@link ANeuralNetworksModel_addOperand}.
*
* @return ANEURALNETWORKS_NO_ERROR if successful.
*/
@@ -2162,7 +2169,7 @@
* Add an operation to a model.
*
* @param model The model to be modified.
- * @param type The {@link ANeuralNetworksOperandType} of the operation.
+ * @param type The {@link ANeuralNetworksOperationType} of the operation.
* @param inputCount The number of entries in the inputs array.
* @param inputs An array of indexes identifying each operand.
* @param outputCount The number of entries in the outputs array.
@@ -2366,7 +2373,9 @@
* model. All other properties of the type must be the
* same as specified in the model. If the type is the same
* as specified when the model was built, NULL can be
- * passed.
+ * passed. Neither the {@link ANeuralNetworksOperandType}
+ * nor the dimensions it points to need to outlive the call
+ * to {@link ANeuralNetworksExecution_setInput}.
* @param buffer The buffer containing the data.
* @param length The length in bytes of the buffer.
*
@@ -2400,7 +2409,9 @@
* to the model. All other properties of the type must be
* the same as specified in the model. If the type is the
* same as specified when the model was built, NULL can be
- * passed.
+ * passed. Neither the {@link ANeuralNetworksOperandType}
+ * nor the dimensions it points to need to outlive the call
+ * to {@link ANeuralNetworksExecution_setInputFromMemory}.
* @param memory The memory containing the data.
* @param offset This specifies the location of the data within the memory.
* The offset is in bytes from the start of memory.
@@ -2437,7 +2448,9 @@
* model. All other properties of the type must be the
* same as specified in the model. If the type is the same
* as specified when the model was built, NULL can be
- * passed.
+ * passed. Neither the {@link ANeuralNetworksOperandType}
+ * nor the dimensions it points to need to outlive the call
+ * to {@link ANeuralNetworksExecution_setOutput}.
* @param buffer The buffer where the data is to be written.
* @param length The length in bytes of the buffer.
*
@@ -2471,7 +2484,9 @@
* model. All other properties of the type must be the
* same as specified in the model. If the type is the same
* as specified when the model was built, NULL can be
- * passed.
+ * passed. Neither the {@link ANeuralNetworksOperandType}
+ * nor the dimensions it points to need to outlive the call
+ * to {@link ANeuralNetworksExecution_setOutputFromMemory}.
* @param memory The memory where the data is to be stored.
* @param offset This specifies the location of the data within the memory.
* The offset is in bytes from the start of memory.
diff --git a/nn/runtime/include/NeuralNetworksOEM.h b/nn/runtime/include/NeuralNetworksOEM.h
index 5995fe9..8218ad9 100644
--- a/nn/runtime/include/NeuralNetworksOEM.h
+++ b/nn/runtime/include/NeuralNetworksOEM.h
@@ -27,6 +27,11 @@
ANEURALNETWORKS_TENSOR_OEM_BYTE = 10001,
}; // extends OperandCode
+/**
+ * If a model contains an {@link ANEURALNETWORKS_OEM_OPERATION}, then
+ * either the model must contain only a single operation, or every
+ * tensor operand type in the model must be fully specified.
+ */
enum {
/** OEM specific operation. */
ANEURALNETWORKS_OEM_OPERATION = 10000,
