HAL interface for compilation and execution hints
The following AIDL types are added:
- TokenValuePair
- PrepareModelConfig
- ExecutionConfig
The following AIDL methods are added:
- IDevice::prepareModelWithConfig
- IPreparedModel::executeSynchronouslyWithConfig
- IPreparedModel::executeFencedWithConfig
- IBurst::executeSynchronouslyWithConfig
The compilation and execution hints are being stored as a list of
token-value pairs as part of the PrepareModelConfig / ExecutionConfig.
And the PrepareModelConfig / ExecutionConfig parcelables are created in
order to make future extensions to the execution related interfaces
easier.
It is the drivers responsibility to verify the hints, and it is allowed
for the driver to ignore them.
Bug: 203248587
Test: neuralnetworks_utils_hal_aidl_test
Change-Id: I98240fd75089fc85cdfcaa0be28aab8a6f0dfca5
Merged-In: I98240fd75089fc85cdfcaa0be28aab8a6f0dfca5
(cherry picked from commit 0e671f3edb9d2c78658a4ef4169e3211e3f9bb00)
diff --git a/neuralnetworks/1.1/utils/src/Device.cpp b/neuralnetworks/1.1/utils/src/Device.cpp
index 3effa84..28f3276 100644
--- a/neuralnetworks/1.1/utils/src/Device.cpp
+++ b/neuralnetworks/1.1/utils/src/Device.cpp
@@ -143,7 +143,9 @@
nn::GeneralResult<nn::SharedPreparedModel> Device::prepareModel(
const nn::Model& model, nn::ExecutionPreference preference, nn::Priority /*priority*/,
nn::OptionalTimePoint /*deadline*/, const std::vector<nn::SharedHandle>& /*modelCache*/,
- const std::vector<nn::SharedHandle>& /*dataCache*/, const nn::CacheToken& /*token*/) const {
+ const std::vector<nn::SharedHandle>& /*dataCache*/, const nn::CacheToken& /*token*/,
+ const std::vector<nn::TokenValuePair>& /*hints*/,
+ const std::vector<nn::ExtensionNameAndPrefix>& /*extensionNameToPrefix*/) const {
// Ensure that model is ready for IPC.
std::optional<nn::Model> maybeModelInShared;
const nn::Model& modelInShared =