Hi Muzamil,

For "any_thread/all_thread', "thread' refers to hardware threads of the
core (like SMT). If you want to count the instructions across all cores,
you will need to add some logic in your python script. Here's an example of
how to do it:
https://github.com/powerjg/gem5/blob/features/tardis/configs/myconfigs/system/system.py#L121
.

Cheers,
Jason

On Wed, Apr 26, 2017 at 1:48 PM Muzamil Rafique <[email protected]>
wrote:

> Hi All,
>
> I am running four different benchmarks on a quad-core CPU. I have two
> questions:
>
> 1. What is the difference between: max_insts_all_threads and
> max_insts_any_thread?
>
> ***
> /*  code from BaseCPU.py  */
>
>  max_insts_all_threads = Param.Counter(0,
>         "terminate when all threads have reached this inst count")
>     max_insts_any_thread = Param.Counter(0,
>         "terminate when any thread reaches this inst count")
>
> ***
>
> 2. If i use any of the above two options, the simulation terminates if
> thread count of application running on "any" of the CPU core reaches that
> limit. What I have to do in order to make sure the simulation terminates
> when "sum" of threads running on all the cores reaches that limit?
>
> Thanks in advance
>
> Muzamil
>
>
> _______________________________________________
> gem5-users mailing list
> [email protected]
> http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
_______________________________________________
gem5-users mailing list
[email protected]
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

Reply via email to