[PATCH 04/10] nohz: Appropriate timekeeper kick on sysidle break

2014-07-28 Thread Frederic Weisbecker
When a CPU wakes up from idle and finds out that the timekeeper is sleeping, we need to kick it such that it switches from dynticks to periodic mode to maintain its timekeeping duty on behalf of the newly awoken CPU. However we aren't using the right API for that. rcu_kick_nohz_cpu() is aimed at

[PATCH 04/10] nohz: Appropriate timekeeper kick on sysidle break

2014-07-28 Thread Frederic Weisbecker
When a CPU wakes up from idle and finds out that the timekeeper is sleeping, we need to kick it such that it switches from dynticks to periodic mode to maintain its timekeeping duty on behalf of the newly awoken CPU. However we aren't using the right API for that. rcu_kick_nohz_cpu() is aimed at

Re: [PATCH 04/10] nohz: Appropriate timekeeper kick on sysidle break

2014-07-21 Thread Paul E. McKenney
On Sat, Jul 19, 2014 at 02:44:15AM +0200, Frederic Weisbecker wrote: > When a CPU wakes up from idle and finds out that the timekeeper is > sleeping, we need to kick it such that it switches from dynticks to > periodic mode to maintain its timekeeping duty on behalf of the newly > awoken CPU. > >

Re: [PATCH 04/10] nohz: Appropriate timekeeper kick on sysidle break

2014-07-21 Thread Paul E. McKenney
On Sat, Jul 19, 2014 at 02:44:15AM +0200, Frederic Weisbecker wrote: When a CPU wakes up from idle and finds out that the timekeeper is sleeping, we need to kick it such that it switches from dynticks to periodic mode to maintain its timekeeping duty on behalf of the newly awoken CPU.

[PATCH 04/10] nohz: Appropriate timekeeper kick on sysidle break

2014-07-18 Thread Frederic Weisbecker
When a CPU wakes up from idle and finds out that the timekeeper is sleeping, we need to kick it such that it switches from dynticks to periodic mode to maintain its timekeeping duty on behalf of the newly awoken CPU. However we aren't using the right API for that. rcu_kick_nohz_cpu() is aimed at

[PATCH 04/10] nohz: Appropriate timekeeper kick on sysidle break

2014-07-18 Thread Frederic Weisbecker
When a CPU wakes up from idle and finds out that the timekeeper is sleeping, we need to kick it such that it switches from dynticks to periodic mode to maintain its timekeeping duty on behalf of the newly awoken CPU. However we aren't using the right API for that. rcu_kick_nohz_cpu() is aimed at