Refine mistakes from adaptaton of NetBSD's hardclock man page to
FreeBSD:
o clarify what usermode means
o clarify how often hardclock is called
o remove Xr callout(9) since that's done elsewhere
Sponsored by: Netflix
Details
Details
- Reviewers
jhb kib mav - Group Reviewers
manpages - Commits
- rG6329ca325e02: hardclock.9: Refine some details
Diff Detail
Diff Detail
- Repository
- rG FreeBSD src repository
- Lint
Lint Not Applicable - Unit
Tests Not Applicable
Event Timeline
| share/man/man9/hardclock.9 | ||
|---|---|---|
| 41–42 | This is too vague as for me. We could specify that it is called hz times per second when system is busy, but at least twice per second when idle. | |
Comment Actions
mav: how does my suggestion sound?
| share/man/man9/hardclock.9 | ||
|---|---|---|
| 41–42 | hardclock is called adaptively based on pending work. The rate ranges from hz times a second for a very busy system, to as few as twice a second for an idle system. When calls at tick boundaries are omitted, cnt reflects an estimate of the number of elapsed ticks since the last call. Over the long haul, the sum of the cnt argument to this function over a second averages hz, however hz(9) has details important for short time scales. To replace everything in this section down to the usermode argument. | |
| share/man/man9/hardclock.9 | ||
|---|---|---|
| 43 | "rate"? | |
| share/man/man9/hardclock.9 | ||
|---|---|---|
| 43 | yes. Unless there's other changes, though, I'll not push an update to fix it (I've fixed it locally) | |