Messages in this thread |  | | Subject | Re: [PATCH v3 0/5] Improve newidle lb cost tracking and early abort | From | Tim Chen <> | Date | Thu, 28 Oct 2021 14:36:08 -0700 |
| |
On Thu, 2021-10-28 at 14:15 +0200, Vincent Guittot wrote: > > > It seems to make sense to skip the call > > to nohz_newidle_balance() for this case? > > nohz_newidle_balance() also tests this condition : > (this_rq->avg_idle < sysctl_sched_migration_cost) > and doesn't set NOHZ_NEWILB_KICKi in such case > > But this patch now used the condition : > this_rq->avg_idle < sd->max_newidle_lb_cost > and sd->max_newidle_lb_cost can be higher than > sysctl_sched_migration_cost > > which means that we can set NOHZ_NEWILB_KICK: > -although we decided to skip newidle loop > -or when we abort because this_rq->avg_idle < curr_cost + sd- > >max_newidle_lb_cost > > This is even more true when sysctl_sched_migration_cost is lowered > which is your case IIRC > > The patch below ensures that we don't set NOHZ_NEWILB_KICK in such > cases: >
Thanks. Will ask our benchmark team to give it a spin.
Tim
|  |