You would run something like

# ./start_sim; bench1 bench2 ;./stop_sim

This website, https://wiki.archlinux.org/index.php/QEMU should have the
information you need to put something on an image.

The ptlsim.log is for a specific run of MARSS. If you use the yamlstats
option you'll see more detailed statistics.

On Thu, Mar 28, 2013 at 11:54 AM, Mansour <[email protected]> wrote:

> Thanks again Tyler.
> The second question's response is very clear. However, I am still a little
> bit
> unsure about comprehending the first one. I run the simulations with the
> "./start_sim; ./microbenchmark; ./stop_sim" command in the guest's
> terminal.
> According to your suggested method, I think must run the following command
> to
> synchronize the beginning time of the execution.
>
> at 12:00
> at > ./start_sim; ./microbenchmark1 > micro_out1; ./stop_sim
> at > ./start_sim; ./microbenchmark2 > micro_out2; ./stop_sim
> at > ./start_sim; ./microbenchmark3 > micro_out3; ./stop_sim
> at > ./start_sim; ./microbenchmark4 > micro_out4; ./stop_sim
> ctrl+d
>
> However, I have some questions about this:
>
> 1) The simulated machine exits after receiving the stop_sim command and I
> see in
> the host's terminal that it shows the message of "MARSSx86::Command
> received :
> -kill". This is not too bothering when I just want to run a single
> micro-benchamark since the guest machine is killed after finishing the
> execution
> of that single micro-benchmark and I can extract the statistics of the
> simulation from the files generated in the host machine. But, this is
> disturbing
> in the case of running multiple micro-benchmarks simultaneously.
>  Obviously,
> finishing the first benchmarks' execution  will cause the guest machine to
> go
> off and the other micro-benchmarks' execution will remain unfinished.
>
> 2) Your provided splash image does not include the "at" command. How can I
> install it on that image? Is there any way to get connected to the
> Internet and
> do sudo apt-get?
>
> 3) How can I differentiate the stat outputs of the simulations? Is there
> any
> facility to print the statistics of each micro-benchmark to a different
> .vis and
> .log files?
>
>
> Thanks in advance for your help
>
>
>
>
>
>
>
> _______________________________________________
> 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