Re: [PATCH 1/3] cpuidle: Store the idle start time stamp

From: Peter Zijlstra
Date: Wed Apr 15 2015 - 09:12:19 EST

On Wed, Apr 15, 2015 at 02:50:33PM +0200, Daniel Lezcano wrote:
> I was thinking about converting to nanosecond the cpuidle framework but it
> is not worth to do that as the resolution is too high for the idle states.

The question is if saving those 4 bytes (unsigned int vs u64) on
next_timer_us is worth having to do that /1000 all the time.

The one spot where its used:

new_factor += RESOLUTION * measured_us / data->next_timer_us;

Could be fixed with a few shifts, all that matters is that measured_us
and next_timer_us are in the same metric, it doesn't need to be us, it
could be ns/1024 for instance.
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at
Please read the FAQ at