Difference sm__cycles_elapsed and smsp__cycles_elapsed?

Accelerated Computing CUDA CUDA Programming and Performance
1

Hello! I have a dumb question :), but I’m a little confused.
I don’t understand clearly the difference between this metrics, and how to the cycles counters related to the HW architecture.

My current understanding (AI100):

  1. Device → SMs->4 sub partitions{An individual: warp scheduler, …, Execution units}.
  2. SM and each sub partition have individual cycle counters.
    This is right?
    Thanks!
2

I recommend asking questions like this on the Nsight Compute forum.

1 Like
3

Oh, sorry, of course, I did Difference sm__cycles_elapsed/smsp__cycles_elapsed and sm__inst_executed/smsp__inst_executed?

4

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.