Call for testers, significant sched_lock patch.

youshi10 at u.washington.edu youshi10 at u.washington.edu
Mon Jun 4 00:43:51 UTC 2007


On Mon, 4 Jun 2007, Abdullah Ibn Hamad Al-Marri wrote:

> On 6/1/07, Jeff Roberson <jroberson at chesapeake.net> wrote:
>> I am going to commit a patch that goes significantly towards decomposing
>> the scheduler lock into per-cpu scheduler locks.  Before this, I would
>> like it if a few more people would run it under more circumstances than I
>> have.  The patch is available at:
>> 
>> http://people.freebsd.org/~jeff/threadlock.diff
>> 
>> So far Kris Kennaway and myself have tested it fairly thoroughly on amd64
>> machines with 1-8 processors.  We've mainly done various benchmarks as
>> well as Peter Holm's stress2 suite.  It has survived our testing so this
>> isn't totally unstable code.  I would like it if more people would test
>> on any other architecture you have available.  I have also tested with
>> 4BSD and ULE with and without INVARIANTS and WITNESS.
>> 
>> This patch should have 0 or almost 0 effect on performance.  It sets the
>> stage for a drop-in replacement for ULE that has per-cpu run queue locks.
>> That scheduler has a sometimes dramatic effect on performance, depending
>> on the workload.  I'm mostly looking for regressions here and not any
>> particular perf impact.
>> 
>> Any questions on the design of the approach should be directed at the
>> arch@ topics on the subject.
>> 
>> Thanks,
>> Jeff
>
> I wish we could try it in i386, do you have plans to do it?
>
>
> -- 
> Regards,
>
> -Abdullah Ibn Hamad Al-Marri
> Arab Portal
> http://www.WeArab.Net/

I'll give it a test on my desktop, once I can verify that the changes to bind and world are sound.

-Garrett



More information about the freebsd-current mailing list