On 05/02/2021 12:48, Vincent Guittot wrote:
> newidle_balance runs with both preempt and irq disabled which prevent
> local irq to run during this period. The duration for updating of the
> blocked load of CPUs varies according to the number of cgroups and
Maybe s/number of cgroups/number of CPU
On Tue, 9 Feb 2021 at 14:09, Valentin Schneider
wrote:
>
> On 05/02/21 12:48, Vincent Guittot wrote:
> > @@ -10517,16 +10499,11 @@ static void nohz_newidle_balance(struct rq
> > *this_rq)
> > time_before(jiffies, READ_ONCE(nohz.next_blocked)))
> > return;
> >
>
> I was
On 05/02/21 12:48, Vincent Guittot wrote:
> @@ -10517,16 +10499,11 @@ static void nohz_newidle_balance(struct rq *this_rq)
> time_before(jiffies, READ_ONCE(nohz.next_blocked)))
> return;
>
I was wondering whether all the conditions above were still relevant. I
think they
newidle_balance runs with both preempt and irq disabled which prevent
local irq to run during this period. The duration for updating of the
blocked load of CPUs varies according to the number of cgroups and
extends this critical period to an uncontrolled level.
Remove the update from
4 matches
Mail list logo