Hi Jeff,
OK, as you know there are 2 sets of 4 functional units (.D, .L, .M, .S) so you can have 8 commands executed in parallel (1 cycle), you can see the commands and what functional unit in sprugh7.pdf -TMS320C66x DSP CPU and Instruction Set in Appendix B table B-1, now there are commands which uses SIMD like the ADD2 (2 additions of 16bit) or ADD4 (4 additions of 8bit) now to use the asm commands use Intrinsics,
Thanks,
HR