Template Function Pennylane::LightningGPU::Util::GEMM_CUDA_device

• Defined in File LinearAlg.hpp

Function Documentation

template<class T = cuDoubleComplex, class DevTypeID = int>
inline void Pennylane::LightningGPU::Util::GEMM_CUDA_device(T *A, T *B, T *C, const int m, const int k, const int n, DevTypeID dev_id, cudaStream_t stream_id, const CublasCaller &cublas)

cuBLAS backed matrix-matrix multiply for GPU data.

Template Parameters

T – Complex data-type. Accepts cuFloatComplex and cuDoubleComplex

Parameters

• A – Device data pointer of matrix A.

• B – Device data pointer of matrix B.

• C – Device data pointer of matrix C.

• m – Row size of the matrix A.

• n – Column size of the matrix B.

• k – Column size of the matrix A and row size of the matrix B.

• dev_id – the device on which the function should be executed.

• stream_id – the CUDA stream on which the operation should be executed.

• cublas – the CublasCaller object that manages the cuBLAS handle.

api/function_LinearAlg_8hpp_1a8bf3a22ae80fe1a48c5c7a5c9b5de064

 

Download Python script

 

Download Notebook

 

View on GitHub