the '-s' option you used, forces gem5 to switch to O3CPU right after
beginning the simulation. So you can see the O3CPU messages. That is
why you thinked '-s' is needed. However that is not a solution.

Normally if you use:
--debug-flags=O3CPUAll <config-file> -F 1000000 --maxinst 1000000 -d
--caches ...

The debug messages are turned on at the beginning of the simulation.
However the first 1M instructions are simulated with AtomicSimpleCPU.
This model has not any pipeline stages. So you see only fetch and
decode messages.

To fix that,you have to find at which "tick" number, the simulation
enters detailed mode (O3CPU). Then pass this number to command line
with "--trace-start". For example, if you see this:

Switched CPUS @ tick 1254044000
1254044000: system.switch_cpus: Switching out
Changing memory mode to timing
switching cpus

Then your command should be:
--debug-flags=O3CPUAll --trace-start=1254044000 <config-file> -F
1000000 --maxinst 1000000 -d --caches ...


Hope that helps

On 2/7/12, Edward Kim <[email protected]> wrote:
> I got it.
> Options "--caches -s" are needed.
> Thanks
>
> --Kim
>
> On Mon, Feb 6, 2012 at 9:08 PM, Gabriel Michael Black <[email protected]
>> wrote:
>
>> Quoting Edward Kim <[email protected]>:
>>
>>
>>>> se.py uses the atomic simple CPU by default, but it can use the other
>>>> ones
>>>> too. Pass it --help to see what options it takes.
>>>>
>>>>
>>>>
>>> I guess the option is as follows..
>>>
>>> -s, --standard-switch
>>>                        switch from timing CPU to Detailed CPU
>>>
>>> But, I have error messages.
>>>
>>> [~/gem5-test]  build/ALPHA_SE/m5.opt configs/example/se.py
>>> --standard-switch O3CPU
>>>
>>> Error: script doesn't take any positional arguments
>>>
>>> Could you please fix it?
>>>
>>>
>> I don't know how exactly --standard-switch is supposed to work, but it
>> doesn't look like it takes any arguments. It's also for running with the
>> timing simple CPU and then switching to O3, not running with O3 to start
>> with. You want -d, I think.
>>
>> Gabe
>>
>>
>> ______________________________**_________________
>> gem5-users mailing list
>> [email protected]
>> http://m5sim.org/cgi-bin/**mailman/listinfo/gem5-users<http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users>
>>
>


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

Reply via email to