SCHED_ULE should not be the default
Steve Kargl
sgk at troutmask.apl.washington.edu
Fri Dec 23 19:11:47 UTC 2011
On Thu, Dec 22, 2011 at 04:23:29PM -0800, Adrian Chadd wrote:
> On 22 December 2011 11:47, Steve Kargl <sgk at troutmask.apl.washington.edu> wrote:
>
> > There is the additional observation in one of my 2008
> > emails (URLs have been posted) that if you have N+1
> > cpu-bound jobs with, say, job0 and job1 ping-ponging
> > on cpu0 (due to ULE's cpu-affinity feature) and if I
> > kill job2 running on cpu1, then neither job0 nor job1
> > will migrate to cpu1. ?So, one now has N cpu-bound
> > jobs running on N-1 cpus.
>
> .. and this sounds like a pretty serious regression. Have you ever
> filed a PR for it?
>
Ah, so goods news! I cannot reproduce this problem that
I saw 3+ years ago on the 4-cpu node, which is currently
running a ULE kernel. When I killed the (N+1)th job,
the N remaining jobs are spread across the N cpus.
One difference between the 2008 tests and today tests is
the number of available cpus. In 2008, I ran the tests
on a node with 8 cpus, while today's test used only a
node with only 4 cpus. If this behavior is a scaling
issue, I can't currently test it. But, today's tests
are certainly encouraging.
--
Steve
More information about the freebsd-stable
mailing list