cvs commit: src/sys/kern sched_ule.c

[Jeff Roberson]

On Sat, 29 Sep 2007, Kevin Oberman wrote:

>> Date: Sat, 29 Sep 2007 23:25:08 -0400
>> From: "Ben Kaduk" <minimarmot at gmail.com>
>> Sender: owner-cvs-all at freebsd.org
>>
>> On 9/29/07, Garance A Drosehn <gad at freebsd.org> wrote:
>>> At 4:39 PM +0000 9/27/07, Jeff Roberson wrote:
>>>>
>>>>   Modified files:
>>>>     sys/kern             sched_ule.c
>>>>   Log:
>>>>    - ...
>>>>    - Assert that we're not trying to compile ULE on an unsupported
>>>>      architecture.  To date, I believe only i386 and amd64 have
>>>>      implemented the third cpu switch argument required.
>>>>
>>>>   Approved by:    re
>>>
>>> Does this mean that I should not switch to ULE on my single-CPU PowerPC
>>> mini-Mac?
>>>
>>
>> I was under the impression that BSD is preferred to ULE for single-processor
>> systems, irregardless of the processor architecture.
>
> YMMV, but ULE seems to generally work better then 4BSD for interactive
> uniprocessor systems. The preferred scheduler for uniprocessor servers
> is less clear, but many test have shown ULE does better for those
> systems in the majority of cases.

I feel it's safe to say desktop behavior on UP is definitely superior.  I 
think there is no significant difference on UP between 4BSD and ULE except 
perhaps in context switching microbenchmarks where ULE falls behind.

>
> While I believe the plan is that 4BSD be in GENERIC in 7.0, but I
> suspect ULE (which may still need optimizing to do in a few areas) will
> soon be the standard scheduler for all 386 and amd64 systems.

I'm not sure if the plan is settled yet, however you're probably right.

>
> Jeff has done quite a job on ULE.

Thanks,
Jeff

> -- 
> R. Kevin Oberman, Network Engineer
> Energy Sciences Network (ESnet)
> Ernest O. Lawrence Berkeley National Laboratory (Berkeley Lab)
> E-mail: oberman at es.net			Phone: +1 510 486-8634
> Key fingerprint:059B 2DDF 031C 9BA3 14A4  EADA 927D EBB3 987B 3751
>