On Tue, Sep 25, 2012 at 1:57 PM, Mehrtash Manoochehri <[email protected]>wrote:

> Hi
> I run parsec without ./start_sim and ./stop_sim and with the following
> parameters.
>

./start_sim and ./stop_sim are not recommanded method for running parsec,
as for 10M instructions, you'll be running the 'parsecmgmt' script only.
 You should run parsec benchmarks from start of ROI regions.

- Avadh


> -machine ooo_2_th -kill-after-run -stopinsns 10M
> Everytime I run, the results are different. say around 10% difference. Is
> it correct? I think it should not be because in this way I will be able to
> test my ideas because I don't understand whether the difference is due to
> the modification or is due to MARSS.
> Furthermore, the number of transaction went to DRAMSim is around 10%
> different which is strange because when we have 10M instruction committed,
> the number of cache misses and memeory accesses should be the same or very
> close.
> Would you please help?
>
> --
> Regards,
> Mehrtash Manoochehri
> Ph.D. Student of Electrical Engineering Department, USC
> http://www-scf.usc.edu/~mmanooch/
>
> _______________________________________________
> http://www.marss86.org
> Marss86-Devel mailing list
> [email protected]
> https://www.cs.binghamton.edu/mailman/listinfo/marss86-devel
>
>
_______________________________________________
http://www.marss86.org
Marss86-Devel mailing list
[email protected]
https://www.cs.binghamton.edu/mailman/listinfo/marss86-devel

Reply via email to