On Thu, May 12, 2016 at 03:31:51AM -0700, tip-bot for Morten Rasmussen wrote: > Commit-ID: cfa10334318d8212d007da8c771187643c9cef35 > Gitweb: http://git.kernel.org/tip/cfa10334318d8212d007da8c771187643c9cef35 > Author: Morten Rasmussen <morten.rasmus...@arm.com> > AuthorDate: Fri, 29 Apr 2016 20:32:40 +0100 > Committer: Ingo Molnar <mi...@kernel.org> > CommitDate: Thu, 12 May 2016 09:55:33 +0200 > > sched/fair: Correct unit of load_above_capacity > > In calculate_imbalance() load_above_capacity currently has the unit > [capacity] while it is used as being [load/capacity]. Not only is it > wrong it also makes it unlikely that load_above_capacity is ever used > as the subsequent code picks the smaller of load_above_capacity and > the avg_load > > This patch ensures that load_above_capacity has the right unit > [load/capacity]. > > Signed-off-by: Morten Rasmussen <morten.rasmus...@arm.com> > [ Changed changelog to note it was in capacity unit; +rebase. ] > Signed-off-by: Peter Zijlstra (Intel) <pet...@infradead.org> > Cc: Dietmar Eggemann <dietmar.eggem...@arm.com> > Cc: Linus Torvalds <torva...@linux-foundation.org> > Cc: Mike Galbraith <efa...@gmx.de> > Cc: Peter Zijlstra <pet...@infradead.org> > Cc: Thomas Gleixner <t...@linutronix.de> > Cc: linux-kernel@vger.kernel.org > Link: > http://lkml.kernel.org/r/1461958364-675-4-git-send-email-dietmar.eggem...@arm.com > Signed-off-by: Ingo Molnar <mi...@kernel.org> > --- > kernel/sched/fair.c | 6 ++++-- > 1 file changed, 4 insertions(+), 2 deletions(-) > > diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c > index 2338105..218f8e8 100644 > --- a/kernel/sched/fair.c > +++ b/kernel/sched/fair.c > @@ -7067,9 +7067,11 @@ static inline void calculate_imbalance(struct lb_env > *env, struct sd_lb_stats *s > if (busiest->group_type == group_overloaded && > local->group_type == group_overloaded) { > load_above_capacity = busiest->sum_nr_running * > SCHED_CAPACITY_SCALE; > - if (load_above_capacity > busiest->group_capacity) > + if (load_above_capacity > busiest->group_capacity) { > load_above_capacity -= busiest->group_capacity; > - else > + load_above_capacity *= NICE_0_LOAD; > + load_above_capacity /= busiest->group_capacity; > + } else > load_above_capacity = ~0UL; > } Hi Morten,
I got the feeling this might be wrong, the NICE_0_LOAD should be scaled down. But I hope I am wrong. Vincent, could you take a look?
