RE: [PATCH 3/3] cpuidle: Using the wake_up_all_idle_cpus() to wake up all idle cpus

[Liu, Chuansheng]

> -----Original Message-----
> From: Daniel Lezcano [mailto:daniel.lezcano@xxxxxxxxxx]
> Sent: Thursday, September 04, 2014 9:02 PM
> To: Liu, Chuansheng; peterz@xxxxxxxxxxxxx; luto@xxxxxxxxxxxxxx;
> rjw@xxxxxxxxxxxxx; mingo@xxxxxxxxxx
> Cc: linux-pm@xxxxxxxxxxxxxxx; linux-kernel@xxxxxxxxxxxxxxx; Liu, Changcheng;
> Wang, Xiaoming; Chakravarty, Souvik K
> Subject: Re: [PATCH 3/3] cpuidle: Using the wake_up_all_idle_cpus() to wake up
> all idle cpus
> 
> On 09/04/2014 09:17 AM, Chuansheng Liu wrote:
> > Currently kick_all_cpus_sync() or smp_call_function() can not
> > break the polling idle cpu immediately.
> >
> > Here using wake_up_all_idle_cpus() which can wake up the polling idle
> > cpu quickly is much helpful for power.
> 
> Acked-by: Daniel Lezcano <daniel.lezcano@xxxxxxxxxx>
> 
> So IIUC, kick_all_cpus_sync is a broken function, right ?
> 
> Wouldn't make sense to replace all calls to kick_all_cpus by
> wake_up_all_idle_cpus ? and remove this broken function ?
My initial patch(V1) indeed do it, but Andy pointed out some callers of kick_all_cpus_sync()
really want the old behavior.

My fault again that do not have the detailed changelog for the patches.

Pasted the comments from Andy:
==
> Currently using smp_call_function() just woke up the corresponding
> cpu, but can not break the polling idle loop.
>
> Here using the new sched API wake_up_if_idle() to implement it.

kick_all_cpus_sync has other callers, and those other callers want the
old behavior.  I think this should be a new function.

--Andy
==