Dear all,

I have figured it out.

When I assign, multiprocesses[i]=bzip2; all the processes are exactly the
same (objective bzip2).

If I make bzip2 a new liveprocess in every loop, then the assigned
processes are also different.

Thanks,


2014-02-16 20:30 GMT-05:00 Yang Zhang <[email protected]>:

> Dear all,
>
> I am trying to let two cpus running the same cpu2006 benchmark at the same
> time.
>
> if i type the command, "./build/X86/gem5.opt ./configs/example/se.py
> --cpu-type=detailed --caches -c
> "../SPEC/benchspec/CPU2006/401.bzip2/exe/bzip2_base.x86_64-m64-gcc42-nn;../SPEC/benchspec/CPU2006/401.bzip2/exe/bzip2_base.x86_64-m64-gcc42-nn"
> -o
> "../SPEC/benchspec/CPU2006/401.bzip2/data/test/input/control;../SPEC/benchspec/CPU2006/401.bzip2/data/test/input/control"
> -n 2 --maxtime=0.002"
>
> from the results i found both of the cpu were running because there were
> data.
>
> However, if i used a modified version of se.py (attached) to run this
> benchmark, every time it seems only the cpu0 is running. i keep all options
> the same. It looks like the same workload is not assigned to cpu1.
>
> But when I specified two different workloads of cpu2006, this script runs
> well!!!!
>
> Can anyone help me? Thanks a lot.
>
> Yang
>
_______________________________________________
gem5-users mailing list
[email protected]
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

Reply via email to