vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

use helper function instead

Browse source
This commit is contained in:
hongruichen 2024-07-10 00:31:13 +08:00
parent a7be0693ba
commit 9add256efe
1 changed files with 2 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

5
ggml/src/ggml-qnn/tensor.hpp
View file

@ -49,8 +49,7 @@ public:
// TODO: set the quantizeParams base on the tensor type
QNN_TENSOR_SET_RANK(_qnn_tensor, qnn::get_ggml_tensor_rank(tensor));
const bool is_npu = device == QNN_BACKEND_NPU;
if (is_npu) {
if (should_use_mem_handle()) {
QNN_TENSOR_SET_MEM_TYPE(_qnn_tensor, QNN_TENSORMEMTYPE_MEMHANDLE);
QNN_TENSOR_SET_MEM_HANDLE(_qnn_tensor, nullptr);
} else {
@ -171,7 +170,7 @@ private:
}
QNN_LOG_INFO("tensor %s: alloc rpcmem(%p) successfully\n", _tensor_name.c_str(), qnn_buffer);
auto error = _qnn_instance->register_rpcmem(qnn_buffer, &_qnn_tensor);
if (error != QNN_SUCCESS) {
QNN_LOG_WARN("register rpc mem failure, %d\n", (int)error);