Page MenuHomeFreeBSD

sched_{4bsd,ule}.4: sync current scheduler sysctls
Needs ReviewPublic

Authored by mchoo on Thu, Sep 24, 11:35 PM.
Tags
None
Referenced Files
F173343275: D60002.diff
Fri, Sep 25, 8:42 AM
F173312289: D60002.id187636.diff
Fri, Sep 25, 4:37 AM
F173312262: D60002.id187635.diff
Fri, Sep 25, 4:37 AM
F173312004: D60002.diff
Fri, Sep 25, 4:35 AM
F173292211: D60002.id187635.diff
Fri, Sep 25, 1:19 AM
F173291964: D60002.id187636.diff
Fri, Sep 25, 1:16 AM
F173291879: D60002.diff
Fri, Sep 25, 1:15 AM
F173288287: D60002.id187636.diff
Fri, Sep 25, 12:30 AM
Subscribers

Details

Reviewers
olce
Group Reviewers
scheduler
manpages
Summary

MFC after: 2 weeks
Sponsored by: FreeBSD Foundation

Diff Detail

Repository
rG FreeBSD src repository
Lint
Lint Skipped
Unit
No Test Coverage
Build Status
Buildable 77294
Build 74177: arc lint + arc unit

Event Timeline

mchoo retitled this revision from sched_4bsd.4: sync current scheduler sysctls to sched_{4bsd,ule}.4: sync current scheduler sysctls.Thu, Sep 24, 11:35 PM
mchoo edited the summary of this revision. (Show Details)

Remove kern.sched.preemption (This is shared between 4BSD and ULE, so it should be in scheduler.4)

Remove kern.sched.preemption (This is shared between 4BSD and ULE, so it should be in scheduler.4)

It's really nice to move it from here in the commit where you're adding that, that way you're not creating a state where that part of the docs is temporarily missing.

It's really nice to move it from here in the commit where you're adding that, that way you're not creating a state where that part of the docs is temporarily missing.

It's better in general, but that does not always happen (see discussion we had in D54881). Minsoo has nothing to do with the situation here, the change was introduced in eb454937a3c0 ("sys: Move ULE sysctls under kern.sched.ule"), so if anyone is to blame, it's kib@ and in part myself as a reviewer (and I did bad when approving the change, on multiple levels).

It's really nice to move it from here in the commit where you're adding that, that way you're not creating a state where that part of the docs is temporarily missing.

Oh, my bad, you were talking about kern.sched.preemption... I agree with that.