from 9.5 documention:
Data is loaded in the sequence that appears in the input file, except when using multidimensional clustering (MDC) tables, partitioned tables, or the anyorder file type modifier. If a particular sequence is desired, sort the data before attempting a load operation. If clustering is required, the data should be sorted on the clustering index prior to loading.
see
Loading data
which means that yes, you need to reorg it. Or I guess you can sort it when you export data from source table so you get pre-sorted data file.