diff options
Diffstat (limited to 'src/backends/neon/workloads')
| -rw-r--r-- | src/backends/neon/workloads/NeonConvolution2dWorkload.cpp | 13 | ||||
| -rw-r--r-- | src/backends/neon/workloads/NeonConvolution2dWorkload.hpp | 9 |
2 files changed, 13 insertions, 9 deletions
diff --git a/src/backends/neon/workloads/NeonConvolution2dWorkload.cpp b/src/backends/neon/workloads/NeonConvolution2dWorkload.cpp index 144baec0ca..83f761158a 100644 --- a/src/backends/neon/workloads/NeonConvolution2dWorkload.cpp +++ b/src/backends/neon/workloads/NeonConvolution2dWorkload.cpp @@ -21,10 +21,11 @@ namespace armnn using namespace armcomputetensorutils; arm_compute::Status NeonConvolution2dWorkloadValidate(const TensorInfo& input, - const TensorInfo& output, - const Convolution2dDescriptor& descriptor, - const TensorInfo& weights, - const Optional<TensorInfo>& biases) + const TensorInfo& output, + const Convolution2dDescriptor& descriptor, + const TensorInfo& weights, + const Optional<TensorInfo>& biases, + bool isFastMathEnabled) { const arm_compute::TensorInfo aclInputInfo = BuildArmComputeTensorInfo(input, descriptor.m_DataLayout); const arm_compute::TensorInfo aclOutputInfo = BuildArmComputeTensorInfo(output, descriptor.m_DataLayout); @@ -52,7 +53,9 @@ arm_compute::Status NeonConvolution2dWorkloadValidate(const TensorInfo& input, &aclOutputInfo, layerInfo, arm_compute::WeightsInfo(), - aclDilationInfo); + aclDilationInfo, + arm_compute::ActivationLayerInfo(), + isFastMathEnabled); } NeonConvolution2dWorkload::NeonConvolution2dWorkload( diff --git a/src/backends/neon/workloads/NeonConvolution2dWorkload.hpp b/src/backends/neon/workloads/NeonConvolution2dWorkload.hpp index 3fb408dbaa..54e08a2042 100644 --- a/src/backends/neon/workloads/NeonConvolution2dWorkload.hpp +++ b/src/backends/neon/workloads/NeonConvolution2dWorkload.hpp @@ -17,10 +17,11 @@ namespace armnn { arm_compute::Status NeonConvolution2dWorkloadValidate(const TensorInfo& input, - const TensorInfo& output, - const Convolution2dDescriptor& descriptor, - const TensorInfo& weights, - const Optional<TensorInfo>& biases); + const TensorInfo& output, + const Convolution2dDescriptor& descriptor, + const TensorInfo& weights, + const Optional<TensorInfo>& biases, + bool isFastMathEnabled = false); class NeonConvolution2dWorkload : public BaseWorkload<Convolution2dQueueDescriptor> { |