AUTH's THMMY "Parallel and distributed systems" course assignments.
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
 
 
 
 
 
 

24 lines
779 B

  1. #!/usr/bin/env bash
  2. #
  3. # prof.sh <exec> <report.file>
  4. #
  5. sudo /usr/local/cuda-11.4/bin/ncu \
  6. --target-processes all \
  7. --metrics "$(echo -n \
  8. "smsp__inst_executed,"\
  9. "smsp__cycles_active.avg,"\
  10. "smsp__cycles_active.sum,"\
  11. "gpu__time_duration.sum,"\
  12. "smsp__average_warp_latency_issue_stalled_barrier,"\
  13. "smsp__warp_issue_stalled_barrier_per_warp_active,"\
  14. "l1tex__average_t_sectors_per_request_pipe_lsu_mem_global_op_ld,"\
  15. "l1tex__average_t_sectors_per_request_pipe_lsu_mem_global_op_st,"\
  16. "l1tex__data_pipe_lsu_wavefronts_mem_shared_cmd_read,"\
  17. "l1tex__data_pipe_lsu_wavefronts_mem_shared_cmd_write,"\
  18. "l1tex__data_bank_conflicts_pipe_lsu_mem_shared_op_ld.sum,"\
  19. "l1tex__data_bank_conflicts_pipe_lsu_mem_shared_op_st.sum "\
  20. )" \
  21. "$1" -q 20 -b 512 > "$2"