Messages in this thread | | | From | Paul Turner <> | Date | Mon, 17 Jun 2013 05:20:46 -0700 | Subject | Re: [patch v8 9/9] sched/tg: remove blocked_load_avg in balance |
| |
On Fri, Jun 7, 2013 at 12:20 AM, Alex Shi <alex.shi@intel.com> wrote: > blocked_load_avg sometime is too heavy and far bigger than runnable load > avg, that make balance make wrong decision. So remove it.
Ok so this is going to have terrible effects on the correctness of shares distribution; I'm fairly opposed to it in its present form.
So let's see, what could be happening..
In "sched: compute runnable load avg in cpu_load and cpu_avg_load_per_task" you already update the load average weights solely based on current runnable load. While this is generally poor for stability (and I suspect the benefit is coming largely from weighted_cpuload() where you do want to use runnable_load_avg and not get_rq_runnable_load() where I suspect including blocked_load_avg() is correct in the longer term).
Ah so.. I have an inkling: Inside weighted_cpuload() where you're trying to use only runnable_load_avg; this is in-fact still including blocked_load_avg for a cgroup since in the cgroup case a group entities' contribution is a function of both runnable and blocked load.
Having weighted_cpuload() pull rq->load (possibly moderated by rq->avg) would reasonably avoid this since issued shares are calculated using instantaneous weights, without breaking the actual model for how much load overall that we believe the group has.
> > Changlong tested this patch, found ltp cgroup stress testing get better > performance: https://lkml.org/lkml/2013/5/23/65 > --- > 3.10-rc1 patch1-7 patch1-8 > duration=764 duration=754 duration=750 > duration=764 duration=754 duration=751 > duration=763 duration=755 duration=751 > > duration means the seconds of testing cost. > --- > > And Jason also tested this patchset on his 8 sockets machine: > https://lkml.org/lkml/2013/5/29/673 > --- > When using a 3.10-rc2 tip kernel with patches 1-8, there was about a 40% > improvement in performance of the workload compared to when using the > vanilla 3.10-rc2 tip kernel with no patches. When using a 3.10-rc2 tip > kernel with just patches 1-7, the performance improvement of the > workload over the vanilla 3.10-rc2 tip kernel was about 25%. > --- > > Signed-off-by: Alex Shi <alex.shi@intel.com> > Tested-by: Changlong Xie <changlongx.xie@intel.com> > Tested-by: Jason Low <jason.low2@hp.com> > --- > kernel/sched/fair.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c > index 3aa1dc0..985d47e 100644 > --- a/kernel/sched/fair.c > +++ b/kernel/sched/fair.c > @@ -1358,7 +1358,7 @@ static inline void __update_cfs_rq_tg_load_contrib(struct cfs_rq *cfs_rq, > struct task_group *tg = cfs_rq->tg; > s64 tg_contrib; > > - tg_contrib = cfs_rq->runnable_load_avg + cfs_rq->blocked_load_avg; > + tg_contrib = cfs_rq->runnable_load_avg; > tg_contrib -= cfs_rq->tg_load_contrib; > > if (force_update || abs64(tg_contrib) > cfs_rq->tg_load_contrib / 8) { > -- > 1.7.12 >
| |