Li, Aubrey wrote:
> Clock system/subsystem is special. They are expected to occur every
> 100 times per second. I think they can't be maken less often.
> The current report is really confused. Because the sum of the causes for
> wakups is far away from the number of "Wakeups-from-idle per second".
> My suggestion is, remove clock/cbe_hres_tick/cyclic_timer report at all,
> or just keep clock event.
> That is just my thougts, ;-)
>   
I understand, and agree this could be confusing. The counts of "software 
waking up" and the counts of the "cpu wakeups" are really looking at two 
different things, although one could reasonably assume that they are 
looking at the same thing (i.e. 1 software wakeup == 1 CPU wakeup), and 
be confused when the counts don't indicate this. Perhaps we should just 
be more explicit (either in the command or the documentation) of what's 
being observed? I'm worried that if we change the probes we look at to 
achieve this consistency (saying that "av_dispatch_softvect" happened 
100 times, rather than clock, cbe_hres_tick(), and cyclic_timer() are 
100/sec cyclics, then it will be harder to track down inefficient cyclic 
subsystem consumers....

-Eric

Reply via email to