Hi all,

I am using atomic cpu and i've set memory latency to 0 and I am only using L1 
cache with hit latency of 1. I am running a simulation for 1 second (1 billion 
cycles).

I am assuming that each instruction or micro-ops (i'm using X86) would take 1 
cycle to complete, is this correct? Because if that's true, then I expect that 
the total number of simulated ops to be 1 billion since I'm simulating for 1 
billion cycles.

However, in the stats file I am seeing something like 800+ million simulated 
micro-ops….can someone explain why there is a difference?

Many Thanks,
Jack
_______________________________________________
gem5-users mailing list
[email protected]
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

Reply via email to