28   int            grid_size  = copy_size / block_size;  in CeedDeviceCopyStrided_Cuda()  local
30   if (block_size * grid_size < copy_size) grid_size += 1;  in CeedDeviceCopyStrided_Cuda()
31   copyStridedK<<<grid_size, block_size>>>(d_array, start, stop, step, d_copy_array);  in CeedDeviceCopyStrided_Cuda()
50   int            grid_size  = vec_size / block_size;  in CeedDeviceSetValue_Cuda()  local
52   if (block_size * grid_size < vec_size) grid_size += 1;  in CeedDeviceSetValue_Cuda()
53   setValueK<<<grid_size, block_size>>>(d_array, length, val);  in CeedDeviceSetValue_Cuda()
74   int            grid_size  = set_size / block_size;  in CeedDeviceSetValueStrided_Cuda()  local
76   if (block_size * grid_size < set_size) grid_size += 1;  in CeedDeviceSetValueStrided_Cuda()
77   setValueStridedK<<<grid_size, block_size>>>(d_array, start, stop, step, val);  in CeedDeviceSetValueStrided_Cuda()
98   int            grid_size  = vec_size / block_size;  in CeedDeviceReciprocal_Cuda()  local
100   if (block_size * grid_size < vec_size) grid_size += 1;  in CeedDeviceReciprocal_Cuda()
101   rcpValueK<<<grid_size, block_size>>>(d_array, length);  in CeedDeviceReciprocal_Cuda()
120   int            grid_size  = vec_size / block_size;  in CeedDeviceScale_Cuda()  local
122   if (block_size * grid_size < vec_size) grid_size += 1;  in CeedDeviceScale_Cuda()
123   scaleValueK<<<grid_size, block_size>>>(x_array, alpha, length);  in CeedDeviceScale_Cuda()
142   int            grid_size  = vec_size / block_size;  in CeedDeviceAXPY_Cuda()  local
144   if (block_size * grid_size < vec_size) grid_size += 1;  in CeedDeviceAXPY_Cuda()
145   axpyValueK<<<grid_size, block_size>>>(y_array, alpha, x_array, length);  in CeedDeviceAXPY_Cuda()
167   int            grid_size  = vec_size / block_size;  in CeedDeviceAXPBY_Cuda()  local
169   if (block_size * grid_size < vec_size) grid_size += 1;  in CeedDeviceAXPBY_Cuda()
170   axpbyValueK<<<grid_size, block_size>>>(y_array, alpha, beta, x_array, length);  in CeedDeviceAXPBY_Cuda()
189   int            grid_size  = vec_size / block_size;  in CeedDevicePointwiseMult_Cuda()  local
191   if (block_size * grid_size < vec_size) grid_size += 1;  in CeedDevicePointwiseMult_Cuda()
192   pointwiseMultValueK<<<grid_size, block_size>>>(w_array, x_array, y_array, length);  in CeedDevicePointwiseMult_Cuda()