Re: ULE nice bugs are fixed.

On Fri, 11 Apr 2003, Ian Freislich wrote:

> Jeff Roberson wrote:
> > Could those of you that have been testing it so far please try it out
> > again and let me know how it feels?
>
> ../../../kern/sched_ule.c: In function `sched_clock':
> ../../../kern/sched_ule.c:927: structure has no member named `ksq_rslice'
>
> $FreeBSD: src/sys/kern/sched_ule.c,v 1.22 2003/04/11 03:47:14 jeff Exp $
>
> Is this the right version of the file?  I don't know enough about
> the scheduler to be sure this is the right fix, but I suspect it
> should be kseq->ksq_rslices if the struct kseq definition is anything
> to go by.
>
> --- sched_ule.c.orig    Fri Apr 11 12:10:56 2003
> +++ sched_ule.c Fri Apr 11 11:45:02 2003
> _at__at_ -924,7 +924,7 _at__at_
>          */
>         ke->ke_slice--;
>  #ifdef SMP
> -       kseq->ksq_rslice--;
> +       kseq->ksq_rslices--;
>  #endif
>
>         if (ke->ke_slice > 0)
>
>
> It compiles after this change, but didn't come back after reboot.
> I'm no where near the machine at the moment, so I don't know why
> it hasn't come up.
>
> Ian

Oh, right, you did the right thing there.  I just didn't have a SMP
machine available for testing the last round of changes.  I'll look into
this.

Cheers,
Jeff