4/3/2023 0 Comments Cuda dim3 gtx 960![]() Printf("Pitched - Time = %f Memory = %i bytes \n", timerGPU. Test_kernel_Pitched_2D >(devPtrPitchedA, devPtrPitchedB, devPtrPitchedC, pitchA, pitchB, pitchC, Nrows, Ncols) ![]() Printf("Non-pitched - Time = %f Memory = %i bytes \n", timerGPU.GetCounter(), Nrows * Ncols * sizeof(float)) Test_kernel_2D >(devPtrA, devPtrB, devPtrC, Nrows, Ncols) GpuErrchk(cudaMemcpy2D(devPtrPitchedC, pitchC, hostPtrC, Ncols * sizeof(float), Ncols*sizeof(float), Nrows, cudaMemcpyHostToDevice)) ĭim3 gridSize(iDivUp(Ncols, BLOCKSIZE_x), iDivUp(Nrows, BLOCKSIZE_y)) ĭim3 blockSize(BLOCKSIZE_y, BLOCKSIZE_x) GpuErrchk(cudaMemcpy2D(devPtrPitchedB, pitchB, hostPtrB, Ncols * sizeof(float), Ncols*sizeof(float), Nrows, cudaMemcpyHostToDevice)) GpuErrchk(cudaMemcpy2D(devPtrPitchedA, pitchA, hostPtrA, Ncols * sizeof(float), Ncols*sizeof(float), Nrows, cudaMemcpyHostToDevice)) GpuErrchk(cudaMallocPitch(
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |