On Thu, January 24, 2013 4:01 pm, Rodrigo Reynolds Ramírez wrote:
>
> I checked the config.ini and the latency in [system.physmem] are
> different. I don't know how to compare the performance of the execution of
> both programs.
>
>> Date: Thu, 24 Jan 2013 13:58:39 -0600
>> Subject: Re: [gem5-users] About statistics
>> From: [email protected]
>> To: [email protected]
>> CC: [email protected]
>>
>> On Thu, January 24, 2013 8:59 am, Rodrigo Reynolds Ramírez wrote:
>> >
>> > Hello everyone:
>> > I am trying to compare different memories architectures, my system is
>> > finally working. But now I have a little problem, I don't see any
>> measure
>> > about performance like time, op/s or something like that in the
>> stats.txt.
>> > Actually I change the main memory latency but I don't see any relevant
>> > change in the results. As well as the miss rate which other measure of
>> > performance could I use?
>>
>> Is the file config.ini from the two runs (before and after changing the
>> latency of the memory) same or different?
>>

Did your simulation actually run? Because the topmost statistic listed in
the file stats.txt, which is sim_seconds, denotes the time spent in the
simulated system for the simulation.

--
Nilay

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

Reply via email to