[E:onnxruntime:, qnn_execution_provider.cc:591 GetCapability] QNN SetupBackend failed qnn_backend_manager.cc:334 InitializeBackend Failed to initialize backend

[June1124]

Describe the issue

yolov8-pose inference using onnxruntime, EP for GPUs, QUALCOMM 8155

To reproduce

std::wstring widestr = std::wstring(model_path.begin(), model_path.end());
qnn_options["backend_path"] = "/data/local/tmp/libQnnGpu.so";//libQnnHtp.so

// qnn_options["profiling_level"] = "basic";//QNN profiling level, options: 'basic', 'detailed', default 'off'.
// qnn_options["htp_performance_mode"] = "sustained_high_performance";//QNN performance mode, options: 'burst', 'balanced', 'default', 'high_performance', 'high_power_saver', 'low_balanced', 'extreme_power_saver', 'low_power_saver', 'power_saver', 'sustained_high_performance'. Default to 'default'.
sessionOptions.AppendExecutionProvider("QNN", qnn_options);

ort_session = new Session(env, model_path.c_str(), sessionOptions);
size_t numInputNodes = ort_session->GetInputCount();
size_t numOutputNodes = ort_session->GetOutputCount();

Urgency

No response

Platform

Android

OS Version

9.0

ONNX Runtime Installation

Built from Source

Compiler Version (if 'Built from Source')

31

Package Name (if 'Released Package')

None

ONNX Runtime Version or Commit ID

1.17.0

ONNX Runtime API

C++/C

Architecture

ARM64

Execution Provider

SNPE, Other / Unknown

Execution Provider Library Version

QNN

[jywu-msft]

Have you tried OnnxRuntime 1.18.0 ?
And which QNN SDK version are you using?

[June1124]

Have you tried OnnxRuntime 1.18.0 ? And which QNN SDK version are you using?

QNN SDK：2.18.0.240101
I confirmed the onnxruntime release notes that onnx 1.17.0 supports qnn 2.18.

[June1124]

Have you tried OnnxRuntime 1.18.0 ? And which QNN SDK version are you using?

Do I need to test in onnxruntime 1.18.0?