Messages in this thread | | | Date | Thu, 1 Apr 2010 10:02:30 -0700 | From | Chris Wright <> | Subject | Re: RFC: Ideal Adaptive Spinning Conditions |
| |
* Darren Hart (dvhltc@us.ibm.com) wrote: > Roland Dreier wrote: >> > I'm looking at some adaptive spinning with futexes as a way to help >> > reduce the dependence on sched_yield() to implement userspace >> > spinlocks. Chris, I included you in the CC after reading your comments >> > regarding sched_yield() at kernel summit and I thought you might be >> > interested. >> >> I think you may have the wrong Chris. Chris Wright wasn't at the most >> recent kernel summit and I do have a vague recollection of Chris Mason >> mentioning that Oracle developers found that sched_yield() performed >> better than futexes + kernel scheduling. > > Oh bother, Sorry Chris! I even talked to Chris Wright about this already > in person and he told me I probably was looking for Chris Mason, and I > still got the last names crossed up!
No worries Darren, I happen to be interested in the topic too ;-) (from KVM point of view, which has a slightly different take)
| |