Re: [patch] new scheduler

Pavel Machek (pavel@atrey.karlin.mff.cuni.cz)
Mon, 10 May 1999 14:12:44 +0200


Hi!

> > This easily doable on linux, too. Even on current scheduler (because
> > it has problems with accounting...):
>
> > PS: It would be nice to modify kernel to be _pessimistic_ about
> > idle. Right now it is so optimistic that you can run boy bad and see
> > 0% cpu utilization when it is in fact 80%. It would be much better to
> > report worse values than to report better values.
>
> i cant seem to be able to reproduce the 'bad' behavior on 2.2.8:
>
> [mingo@moon lo]$ time ./badguy
> I'm a good boy.
> ..........4.66 user 0.41 system 0:05.07 elapsed 99%CPU
> [mingo@moon lo]$ time ./badguy bad
> I'm a *BAD* boy! And bad boys know that jiffie is 20msec
> ..........4.66 user 0.43 system 0:05.12 elapsed 99%CPU
> [mingo@moon lo]$

Uch, ouch. I posted code without trying it first. I know it worked
once some time ago, but I was unable to reproduce it. (You always
needed to tune constants "right", but I can not seem to do that now.)

> (i've tried it both on SMP and UP boxes) In both cases the kernel says
> we've spent almost 100% as user-time. (are you sure 2.2 still has this
> problem? I saw your(?) timer tick accounting fix half a year ago, and i
> thought it went into the kernel, but i could be wrong.)

I do not think my patches went in. If it currently works, problem
might have just gone away :-).

Pavel

-- 
The best software in life is free (not shareware)!		Pavel
GCM d? s-: !g p?:+ au- a--@ w+ v- C++@ UL+++ L++ N++ E++ W--- M- Y- R+

- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/