Re: Periodic rant about SCHED_ULE

From: Zaphod Beeblebrox <zbeeble_at_gmail.com>
Date: Tue, 13 Jul 2021 22:24:15 UTC
Also, TBH, I think the goals, proof that algorithm meets the goals and then
code is one of the shining examples of FreeBSD engineering.  It's one of
the things I point to when people ask why I like FreeBSD (after the licence
and the governance)... and thats in sched_4bsd.c.

On Tue, Jul 13, 2021 at 6:21 PM George Mitchell <george+freebsd@m5p.com>
wrote:

> On 7/13/21 6:09 PM, Zaphod Beeblebrox wrote:
> > [...]
> > TBH, I think SCHED_ULE is a failure and the only reason more people don't
> > think so is that processors are now largely too fast for people to care.
> > [...]
> I won't call it a failure, but I will call it the wrong default for the
> GENERIC build.  And you're certainly right about why people don't care.
> -- George
> >
> > I know we don't have guiding principles for nice, but I would toss out
> the
> > +/- five rule for it --- that any process more than 5 nice levels lower
> > from a cpu-busy process shouldn't preempt the higher process.  I realize
> we
> > have rtprio, but it's a pain to use.  Anyways, don't let this last
> comment
> > distract.
> > [...]
>
>