Re: pps_capture() and pps_fetch()

From: Poul-Henning Kamp <phk_at_phk.freebsd.dk>
Date: Wed, 08 Jun 2022 08:25:49 UTC
--------
Sebastian Huber writes:

> 		tc_min_ticktock_freq =3D max(1, tc->tc_frequency /
> 		    (((uint64_t)tc->tc_counter_mask + 1) / 3));
>
> code in tc_windup(). I don't understand the value "3" here, I guess it
> should depend on timehands_count?

Not really.

The divide by three is holding a good distance to the half-way mark, so
that we "wind" to the next hand before the roll-over introduces s
ambiguity.

The number of timehands relate to how bad latency (interrupt and
scheduling) can be for kernel threads.

-- 
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk@FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe    
Never attribute to malice what can adequately be explained by incompetence.