svn commit: r215281 - head/sys/kern

[Bruce Evans]

On Sun, 14 Nov 2010, Bruce Cran wrote:

> Log:
>  Add some descriptions to sys/kern sysctls.

> Modified: head/sys/kern/kern_tc.c
> ==============================================================================
> --- head/sys/kern/kern_tc.c	Sun Nov 14 05:05:41 2010	(r215280)
> +++ head/sys/kern/kern_tc.c	Sun Nov 14 06:09:50 2010	(r215281)
> @@ -103,7 +103,7 @@ SYSCTL_NODE(_kern_timecounter, OID_AUTO,
>
> static int timestepwarnings;
> SYSCTL_INT(_kern_timecounter, OID_AUTO, stepwarnings, CTLFLAG_RW,
> -    &timestepwarnings, 0, "");
> +    &timestepwarnings, 0, "Log time steps");

Another problem (other than style bugs) with these and some old descriptions
is inconsistent descriptions of boolean variables.  I like always starting
with "Enable".

> @@ -768,7 +768,7 @@ pps_event(struct pps_state *pps, int eve
>  */
>
> static int tc_tick;
> -SYSCTL_INT(_kern_timecounter, OID_AUTO, tick, CTLFLAG_RD, &tc_tick, 0, "");
> +SYSCTL_INT(_kern_timecounter, OID_AUTO, tick, CTLFLAG_RD, &tc_tick, 0, "Approximate number of hardclock ticks in a millisecond");

It can't possibly be that if a hardclock tick is much larger than a
millisecond, and this doesn't describe its purpose.  It is actually a
rate limiter for timecounter updates, to avoid the natural rate of once
per hardclock tick if that is considered too frequent.  Thus it has nothing
to do with milliseconds except once per millisecond is considered frequent
enough.  It is the number of hardclock ticks between timercounter updates.

Bruce