conjugateGradientMultiDeviceCG -dc cudaMemAdvise cudaMemPrefetchAsync cudaLaunchCooperativeKernelMultiDevice cudaStreamSynchronize cudaOccupancyMaxActiveBlocksPerMultiprocessor whole ./ ../ ../../common/inc Unified Memory Linear Algebra Cooperative Groups MultiDevice Cooperative Groups CUDA Sparse Matrix Unified Memory Multi-GPU cudadevrt true conjugateGradientMultiBlockCG.cu UVM MDCG 1:CUDA Advanced Topics 3:Linear Algebra sm60 sm61 sm70 x86_64 linux ppc64le linux 6.0 exe