On Thu, Sep 15, 2016 at 06:36:53PM +0100, Dietmar Eggemann wrote: > > We did however loose a lot on why and how min(1, runnable_avg) is a > > sensible thing to do... > > Do you refer to the big comment on top of this if condition in the old > code in __update_group_entity_contrib()? The last two subsections of it > I never understood ... > > [...]
Read the thread here: http://www.gossamer-threads.com/lists/linux/kernel/1558210 But Vince's explanation earlier is also good.