On Tue, Nov 2, 2010 at 7:33 AM, Weixun Wang <[email protected]> wrote:

> Hi all,
>
> A simple question. Could any one tell me that is the miss latency (e.g.,
> system.cpu.dcache.overall_miss_latency) in M5 simulation output file? It is
> apparently not "# number of overall miss cycles" since it has a very huge
> value (e.g., 751856570000 while the cpu cycles simulated is only 783263851).
> Moreover, how is this latency calculated?
>

The overall miss cycles is the sum of the latencies of all the misses; you
divide by the number of misses to get the average miss latency.


>
> And I'm assuming all cache access latency, miss latency, block fill time
> are incorporated in "system.cpu.numCycles", right?
>

I don't remember exactly what cpu.numCycles tracks vs. sim_ticks (I think it
excludes idle time), but it definitely includes memory access latencies.


>
> Thanks.
>
> --
> Best Regards,
>
> Wang, Weixun
>
> _______________________________________________
> m5-users mailing list
> [email protected]
> http://m5sim.org/cgi-bin/mailman/listinfo/m5-users
>
_______________________________________________
m5-users mailing list
[email protected]
http://m5sim.org/cgi-bin/mailman/listinfo/m5-users

Reply via email to