diff options
author | Colm Donelan <colm.donelan@arm.com> | 2022-07-06 12:09:05 +0100 |
---|---|---|
committer | Nikhil Raj <nikhil.raj@arm.com> | 2022-07-27 15:56:33 +0100 |
commit | d7ceec59ce45f690deba2c0d452ec91fabbdadf9 (patch) | |
tree | ddea3d4092d1d9df21a751bf8cf1750c746ad644 /include/armnn/IRuntime.hpp | |
parent | 9d9dd223ba9fbc509ea8ff1c211d3c63943a5989 (diff) | |
download | armnn-d7ceec59ce45f690deba2c0d452ec91fabbdadf9.tar.gz |
IVGCVSW-6896 Fix pre-import when using sync execute.
* Refactor backend capability checks in LoadedNetwork.
* ImportInputs should check the number of tensors does not exceed the
number of inputs.
* In EnqueueWorkload the check for for the count of input tensors
was ignoring pre-imported inputs.
* Added checks to verify ImportInputs/ImportOutputs worked as expected
in EndToEndTestImpl.
* Improve documentation on ImportInputs/ImportOutputs in IRuntime.hpp.
* Disabled import tests in CL and Neon EndToEndTests that cannot work.
Signed-off-by: Colm Donelan <colm.donelan@arm.com>
Change-Id: Iae4b2644a1c9f01ee72bce1afb211661cc9ae2e3
Diffstat (limited to 'include/armnn/IRuntime.hpp')
-rw-r--r-- | include/armnn/IRuntime.hpp | 8 |
1 files changed, 6 insertions, 2 deletions
diff --git a/include/armnn/IRuntime.hpp b/include/armnn/IRuntime.hpp index 0a2dc3233e..9ea9ac83dc 100644 --- a/include/armnn/IRuntime.hpp +++ b/include/armnn/IRuntime.hpp @@ -221,14 +221,18 @@ public: /// ImportInputs separates the importing and mapping of InputTensors from network execution. /// Allowing for a set of InputTensors to be imported and mapped once, but used in execution many times. /// This function is not thread safe and must not be used while other threads are calling Execute(). - /// Only compatible with AsyncEnabled networks and aligned memory import + /// No exceptions are thrown for failed imports. It is the caller's responsibility to check whether + /// tensors have been successfully imported by comparing returned ids with those passed in the InputTensors. + /// Whether a tensor can be imported or not is backend specific. std::vector<ImportedInputId> ImportInputs(NetworkId networkId, const InputTensors& inputTensors, MemorySource forceImportMemorySource = MemorySource::Undefined); /// ImportOutputs separates the importing and mapping of OutputTensors from network execution. /// Allowing for a set of OutputTensors to be imported and mapped once, but used in execution many times. /// This function is not thread safe and must not be used while other threads are calling Execute(). - /// Only compatible with AsyncEnabled networks and aligned memory import + /// No exceptions are thrown for failed imports. It is the caller's responsibility to check whether + /// tensors have been successfully imported by comparing returned ids with those passed in the OutputTensors. + /// Whether a tensor can be imported or not is backend specific. std::vector<ImportedOutputId> ImportOutputs(NetworkId networkId, const OutputTensors& outputTensors, MemorySource forceImportMemorySource = MemorySource::Undefined); |