[PATCH] cpuidle: Fix last_residency division

Nicolas Pitre nicolas.pitre at linaro.org
Fri Jul 1 01:28:14 AEST 2016


On Thu, 30 Jun 2016, Shreyas B. Prabhu wrote:

> Snooze is a poll idle state in powernv and pseries platforms. Snooze
> has a timeout so that if a cpu stays in snooze for more than target
> residency of the next available idle state, then it would exit thereby
> giving chance to the cpuidle governor to re-evaluate and
> promote the cpu to a deeper idle state. Therefore whenever snooze exits
> due to this timeout, its last_residency will be target_residency of next
> deeper state.
> 
> commit e93e59ce5b85 ("cpuidle: Replace ktime_get() with local_clock()")
> changed the math around last_residency calculation. Specifically, while
> converting last_residency value from nanoseconds to microseconds it does
> right shift by 10. Due to this, in snooze timeout exit scenarios
> last_residency calculated is roughly 2.3% less than target_residency of
> next available state. This pattern is picked up get_typical_interval()
> in the menu governor and therefore expected_interval in menu_select() is
> frequently less than the target_residency of any state but snooze.
> 
> Due to this we are entering snooze at a higher rate, thereby affecting
> the single thread performance.
> 
> Fix this by using a better approximation for division by 1000.
> 
> Reported-by: Anton Blanchard <anton at samba.org>
> Bisected-by: Shilpasri G Bhat <shilpa.bhat at linux.vnet.ibm.com>
> Suggested-by David Laight <david.laight at aculab.com>
> Signed-off-by: Shreyas B. Prabhu <shreyas at linux.vnet.ibm.com>

Minor nit:

> +/*
> + * To ensure that there is no overflow while approximation
> + * for dividing val by 1000, we must respect -
> + * val + (val >> 5) <= 0xFFFFFFFF
> + * val + val/32 <= 0xFFFFFFFF
> + * val <= (0xFFFFFFFF * 32) / 33
> + * val <= 0xF83E0F82
> + * Hence the threshold for val below which we can use the
> + * approximation is 0xF83E0F82
> + */
> +#define DIV_APPROXIMATION_THRESHOLD 0xF83E0F82UL
> +
> +/*
> + * Used for calculating last_residency in usec. Optimized for case
> + * where last_residency in nsecs is < DIV_APPROXIMATION_THRESHOLD
> + * Approximated value has less than 1% error.
> + */
> +static inline int convert_nsec_to_usec(u64 nsec)
> +{
> +	if (likely(nsec < DIV_APPROXIMATION_THRESHOLD)) {

To be coherent with the comment, you could use <= instead.

Then you may add:

Reviewed-by: Nicolas Pitre <nico at linaro.org>


Nicolas


More information about the Linuxppc-dev mailing list