Re: [PATCH] ftrace: Add a C/P state tracer to help poweroptimization

[Ingo Molnar]

* Frédéric Weisbecker <fweisbec@xxxxxxxxx> wrote:

> That looks good...

[...]
> 
> Just for curiosity, do you have an example of trace and/or graph 
> from this tracer?

Output looks like this:

# tracer: power
#
#           TASK-PID    CPU#    TIMESTAMP  FUNCTION
#              | |       |          |         |
[   54.698035123] CSTATE: Going to C1 on cpu 0 for 0.000022629
[   54.698059987] CSTATE: Going to C1 on cpu 0 for 0.000074032
[   54.698191011] CSTATE: Going to C1 on cpu 0 for 0.000134655
[   54.698327901] CSTATE: Going to C1 on cpu 0 for 0.000276017
[   54.699178020] CSTATE: Going to C1 on cpu 8 for 0.001454114
[   54.698138489] CSTATE: Going to C1 on cpu 5 for 0.002507334
[   54.699079124] CSTATE: Going to C1 on cpu 4 for 0.001611677
[   54.698605873] CSTATE: Going to C1 on cpu 0 for 0.003998045
[   54.700276498] CSTATE: Going to C1 on cpu 10 for 0.003994971
[   54.700636045] CSTATE: Going to C1 on cpu 6 for 0.003993295
[   54.700637442] CSTATE: Going to C1 on cpu 2 for 0.003993295
[   54.700634648] CSTATE: Going to C1 on cpu 8 for 0.003998883

	Ingo
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/