Hello Everyone ,
I have set up anup arm full system simulation and I am trying to run
benchmarks,the problem I am facing is that it is very slow ,does not reach
the desired part of the benchmark/output . I would like to know how can i
limit the number of instructions executed, indicate the number of
instructions for system warmup and also limit the time of entire
simulation, i see that these options are not available as command line
arguments in the gem5 full system simulation.
I am using arm big little configuration (as i am using fs_power.py) so,
would I have to limit the number of instructions for both the big cluster
and little cluster? is it the same for warmup instructions ?
The fs_power.py also does not allow switching CPU type to Atomicsimple CPU
as it requires TimimgCPU as a compulsory .
Please let me know if there are any other ways also to accelerate my
simulation
Regards
Saras
Why my mail is not being sent?
On Tue, Dec 26, 2023, 19:21 saras nanda via gem5-users gem5-users@gem5.org
wrote:
Hello Everyone ,
I have set up anup arm full system simulation and I am trying to run
benchmarks,the problem I am facing is that it is very slow ,does not reach
the desired part of the benchmark/output . I would like to know how can i
limit the number of instructions executed, indicate the number of
instructions for system warmup and also limit the time of entire
simulation, i see that these options are not available as command line
arguments in the gem5 full system simulation.
I am using arm big little configuration (as i am using fs_power.py) so,
would I have to limit the number of instructions for both the big cluster
and little cluster? is it the same for warmup instructions ?
The fs_power.py also does not allow switching CPU type to Atomicsimple CPU
as it requires TimimgCPU as a compulsory .
Please let me know if there are any other ways also to accelerate my
simulation
Regards
Saras
gem5-users mailing list -- gem5-users@gem5.org
To unsubscribe send an email to gem5-users-leave@gem5.org