Greetings,
I'm taking some measurements on an image without using checkpoints - I
have it set up so that when the image is run and boots, it immediately
logs in to root, drops into simulation, and runs my benchmark. I've
noticed that, when I do this on different machines, the total cycle
count (as seen in stats.yaml and reading the sim_cycle variable) over
the course of simulation varies fairly dramatically depending on which
machine it's on. Since cycles elapsed affects some of my measurements,
I'd like to have the experiment be repeatable across different machines,
and have some assurance that the cycle count I'm seeing is at least
relatively accurate, if possible. Is this variation expected, due to the
way MARSS is set up?
Thanks in advance.
Sincerely,
Addison
_______________________________________________
http://www.marss86.org
Marss86-Devel mailing list
[email protected]
https://www.cs.binghamton.edu/mailman/listinfo/marss86-devel
