inlinePTX cudaMemcpy cudaFree cudaMallocHost cudaGetLastError cudaGridSize cudaBlockSize cudaDeviceSynchronize cudaFreeHost cudaMalloc whole ./ ../ ../../../Common Performance Strategies PTX Assembly CUDA Driver API CUDA true inlinePTX.cu 1:CUDA Basic Topics 1:Performance Strategies sm50 sm52 sm53 sm60 sm61 sm70 sm72 sm75 sm80 sm86 sm87 sm90 x86_64 linux windows7 x86_64 macosx arm sbsa ppc64le linux all exe