darknet源碼解析：cuda_push_array

更多darknet源代碼學習筆記，參看：darknet源碼學習：預測分類函數float *network_predict_gpu(network net, float *input)

將主機host的數據拷貝到GPU設備x_gpu中

void cuda_push_array(float *x_gpu, float *x, size_t n)
{
    size_t size = sizeof(float)*n;
    //cudaError_t status = cudaMemcpy(x_gpu, x, size, cudaMemcpyHostToDevice);
    cudaError_t status = cudaMemcpyAsync(x_gpu, x, size, cudaMemcpyHostToDevice, get_cuda_stream());
    CHECK_CUDA(status);
}

cudaMemcpy與cudaMemcpyAsync的區別

CudaDeviceSynchronize vs cudaThreadSynchronize vs cudaStreamSynchronize