Hi,
I have a simple program for matrix multiplication using OpenVINO (Version: 2024.4.0) on Windows 11. It works fine on CPU and GPU, but it crashes on NPU upon exit. If anyone has any clues on how to fix this, I would really appreciate it. Thank you!
Regards,
-yanny
#include <openvino/openvino.hpp>
#include <openvino/op/matmul.hpp>
#include <iostream>
int main() {
try
{
ov::Core core;
ov::Shape input_shape_A{ 1, 1024 };
ov::Shape input_shape_B{ 1024, 1024 };
auto input_A = std::make_shared<ov::op::v0::Parameter>(ov::element::f32, input_shape_A);
auto input_B = std::make_shared<ov::op::v0::Parameter>(ov::element::f32, input_shape_B);
bool transpose_A = false;
bool transpose_B = true;
auto matmul_op = std::make_shared<ov::op::v0::MatMul>(input_A, input_B, transpose_A, transpose_B);
auto result = std::make_shared<ov::op::v0::Result>(matmul_op);
ov::ParameterVector inputs{ input_A, input_B };
ov::ResultVector results{ result };
auto model = std::make_shared<ov::Model>(results, inputs, "MatMulModel");
// This works on CPU and GPU
auto compiled_model = core.compile_model(model, "NPU");
auto infer_request = compiled_model.create_infer_request();
std::vector<float> input_data_A(1 * 1024, 2.0f);
std::vector<float> input_data_B(1024 * 1024, 1.0f);
infer_request.set_tensor(input_A, ov::Tensor(ov::element::f32, input_shape_A, input_data_A.data()));
infer_request.set_tensor(input_B, ov::Tensor(ov::element::f32, input_shape_B, input_data_B.data()));
infer_request.infer();
infer_request.wait();
auto output_tensor = infer_request.get_output_tensor();
const float* output_data = output_tensor.data<float>();
std::cout << "MatMul result: ";
for (size_t i = 0; i < 1024; ++i) {
std::cout << output_data[i] << " ";
}
std::cout << std::endl;
}
catch (std::exception e)
{
std::cout << "Exception: " << e.what() << std::endl;
}
return 0;
}
