Aubrey Li wrote:
> Hi Rafael,
>
> 2008/3/31 Rafael Vanoni <Rafael.Vanoni at sun.com>:
>   
>> Hey everyone
>>
>>  Here's an export with a lot of code cleanup and fixes for most of what's
>>  on the PowerTop Testing page. Let me know what you think of it, Pat,
>>  once you get around to it.
>>     
>
> Thanks for your work, I really appreciate it!
> The patch looks good, please feel free to commit into repo.
>
>   
>>  I'm also working on the man page, which leads me to a question about the
>>  bug that Aubrey reported. Did you guys reach a decision whether it
>>  should be dealt with in the code or is it something that should just be
>>  well documented ?
>>     
>
> I still think it would be better to deal with the bug in the code.
> something like
>
> 17.8%(100)  <kernel>: unix`cbe_hres_tick, genunix`clock, genunix`cyclic_timer
>
> But I'll defer to Eric, ;-)
>   
I guess there's two ways of looking at this:
A) PowerTOP is reporting on the events that are causing CPUs to wake up.
B) PowerTOP is reporting on the events that are (or can) cause CPUs to 
wake up.

Perhaps one could be the default behavior, and the other could be 
enabled with an option. My impression is that A is the default that 
folks would expect, but I think B is useful. I think B is the way the 
tool works now.

To get A, I think PowerTOP needs "D" code that is able to correlate 
firing of the cyclic-fire probe with a firing of the 
idle-state-transition, and reports only the cyclic-firing that actually 
caused the CPU to wake up.

-Eric


Reply via email to