Re: [Query]: tick-sched: can idle_active be false in tick_nohz_idle_exit()?

2014-04-11 Thread Frederic Weisbecker
On Fri, Apr 11, 2014 at 03:24:11PM +0530, Viresh Kumar wrote: > On 10 April 2014 20:26, Frederic Weisbecker wrote: > > When a dynticks idle CPU is woken up (typically with an IPI), > > tick_nohz_stop_idle() > > is called on interrupt entry but, because this is a waking up IPI, > > tick_nohz_star

Re: [Query]: tick-sched: can idle_active be false in tick_nohz_idle_exit()?

2014-04-11 Thread Viresh Kumar
On 10 April 2014 20:26, Frederic Weisbecker wrote: > When a dynticks idle CPU is woken up (typically with an IPI), > tick_nohz_stop_idle() > is called on interrupt entry but, because this is a waking up IPI, > tick_nohz_start_idle() > won't be called. The reason is that need_resched() prevents

Re: [Query]: tick-sched: can idle_active be false in tick_nohz_idle_exit()?

2014-04-10 Thread Frederic Weisbecker
On Wed, Apr 09, 2014 at 05:28:57PM +0530, Viresh Kumar wrote: > Hi Guys, > > File: kernel/time/tick-sched.c > function: tick_nohz_idle_exit() > > We are checking here if idle_active is true or not and then > do some stuff. But is it possible that idle_active be false > here? > > The sequence as

[Query]: tick-sched: can idle_active be false in tick_nohz_idle_exit()?

2014-04-09 Thread Viresh Kumar
Hi Guys, File: kernel/time/tick-sched.c function: tick_nohz_idle_exit() We are checking here if idle_active is true or not and then do some stuff. But is it possible that idle_active be false here? The sequence as far as I understood is: idle-loop: tick_nohz_idle_enter(), i.e. idle_active = tru