75 …elem, int blocks_per_sm, int max_threads_per_block, int max_threads_z, int warp_size, int block[3],  in BlockGridCalculate()
78   const int threads_per_elem = block[0] * block[1];  in BlockGridCalculate()
94   block[2] = CeedIntMin(elems_per_block, max_threads_z);  in BlockGridCalculate()
209   int block[3] = {data->thread_1d, ((!is_tensor || data->dim == 1) ? 1 : data->thread_1d), -1};  in CeedOperatorApplyAddCore_Cuda_gen()  local
213 …          cuda_data->device_prop.maxThreadsDim[2], cuda_data->device_prop.warpSize, block, &grid));  in CeedOperatorApplyAddCore_Cuda_gen()
218     block[2] = elems_per_block;  in CeedOperatorApplyAddCore_Cuda_gen()
220   CeedInt shared_mem = block[0] * block[1] * block[2] * sizeof(CeedScalar);  in CeedOperatorApplyAddCore_Cuda_gen()
222 …nd(CeedTryRunKernelDimShared_Cuda(ceed, data->op, stream, grid, block[0], block[1], block[2], shar…  in CeedOperatorApplyAddCore_Cuda_gen()
464     int block[3] = {data->thread_1d, ((!is_tensor || data->dim == 1) ? 1 : data->thread_1d), -1};  in CeedOperatorLinearAssembleQFunctionCore_Cuda_gen()  local
468 …          cuda_data->device_prop.maxThreadsDim[2], cuda_data->device_prop.warpSize, block, &grid));  in CeedOperatorLinearAssembleQFunctionCore_Cuda_gen()
473       block[2] = elems_per_block;  in CeedOperatorLinearAssembleQFunctionCore_Cuda_gen()
475     CeedInt shared_mem = block[0] * block[1] * block[2] * sizeof(CeedScalar);  in CeedOperatorLinearAssembleQFunctionCore_Cuda_gen()
477 …ernelDimShared_Cuda(ceed, data->assemble_qfunction, NULL, grid, block[0], block[1], block[2], shar…  in CeedOperatorLinearAssembleQFunctionCore_Cuda_gen()
639     int block[3] = {data->thread_1d, (data->dim == 1 ? 1 : data->thread_1d), -1};  in CeedOperatorLinearAssembleAddDiagonalAtPoints_Cuda_gen()  local
642 …          cuda_data->device_prop.maxThreadsDim[2], cuda_data->device_prop.warpSize, block, &grid));  in CeedOperatorLinearAssembleAddDiagonalAtPoints_Cuda_gen()
643     CeedInt shared_mem = block[0] * block[1] * block[2] * sizeof(CeedScalar);  in CeedOperatorLinearAssembleAddDiagonalAtPoints_Cuda_gen()
645 …KernelDimShared_Cuda(ceed, data->assemble_diagonal, NULL, grid, block[0], block[1], block[2], shar…  in CeedOperatorLinearAssembleAddDiagonalAtPoints_Cuda_gen()
804     int block[3] = {data->thread_1d, (data->dim == 1 ? 1 : data->thread_1d), -1};  in CeedOperatorAssembleSingleAtPoints_Cuda_gen()  local
807 …          cuda_data->device_prop.maxThreadsDim[2], cuda_data->device_prop.warpSize, block, &grid));  in CeedOperatorAssembleSingleAtPoints_Cuda_gen()
808     CeedInt shared_mem = block[0] * block[1] * block[2] * sizeof(CeedScalar);  in CeedOperatorAssembleSingleAtPoints_Cuda_gen()
810 …yRunKernelDimShared_Cuda(ceed, data->assemble_full, NULL, grid, block[0], block[1], block[2], shar…  in CeedOperatorAssembleSingleAtPoints_Cuda_gen()