Re: [PATCH V2] cpuidle: Change ktime_get() with local_clock()

2016-04-27 Thread Rafael J. Wysocki
On Friday, April 22, 2016 08:42:40 AM Peter Zijlstra wrote: > On Thu, Apr 21, 2016 at 09:41:14PM +0200, Rafael J. Wysocki wrote: > > On Thu, Apr 21, 2016 at 10:56 AM, Daniel Lezcano > > wrote: > > > The ktime_get() can have a non negligeable overhead, use local_clock()

Re: [PATCH V2] cpuidle: Change ktime_get() with local_clock()

2016-04-27 Thread Rafael J. Wysocki
On Friday, April 22, 2016 08:42:40 AM Peter Zijlstra wrote: > On Thu, Apr 21, 2016 at 09:41:14PM +0200, Rafael J. Wysocki wrote: > > On Thu, Apr 21, 2016 at 10:56 AM, Daniel Lezcano > > wrote: > > > The ktime_get() can have a non negligeable overhead, use local_clock() > > > instead. > > > > > >

Re: [PATCH V2] cpuidle: Change ktime_get() with local_clock()

2016-04-22 Thread Peter Zijlstra
On Thu, Apr 21, 2016 at 09:41:14PM +0200, Rafael J. Wysocki wrote: > On Thu, Apr 21, 2016 at 10:56 AM, Daniel Lezcano > wrote: > > The ktime_get() can have a non negligeable overhead, use local_clock() > > instead. > > > > In order to test the difference between

Re: [PATCH V2] cpuidle: Change ktime_get() with local_clock()

2016-04-22 Thread Peter Zijlstra
On Thu, Apr 21, 2016 at 09:41:14PM +0200, Rafael J. Wysocki wrote: > On Thu, Apr 21, 2016 at 10:56 AM, Daniel Lezcano > wrote: > > The ktime_get() can have a non negligeable overhead, use local_clock() > > instead. > > > > In order to test the difference between ktime_get() and local_clock(), > >

Re: [PATCH V2] cpuidle: Change ktime_get() with local_clock()

2016-04-21 Thread Rafael J. Wysocki
On Thu, Apr 21, 2016 at 10:56 AM, Daniel Lezcano wrote: > The ktime_get() can have a non negligeable overhead, use local_clock() > instead. > > In order to test the difference between ktime_get() and local_clock(), > a quick hack has been added to trigger, via debugfs,

Re: [PATCH V2] cpuidle: Change ktime_get() with local_clock()

2016-04-21 Thread Rafael J. Wysocki
On Thu, Apr 21, 2016 at 10:56 AM, Daniel Lezcano wrote: > The ktime_get() can have a non negligeable overhead, use local_clock() > instead. > > In order to test the difference between ktime_get() and local_clock(), > a quick hack has been added to trigger, via debugfs, 1 times a > call to

[PATCH V2] cpuidle: Change ktime_get() with local_clock()

2016-04-21 Thread Daniel Lezcano
The ktime_get() can have a non negligeable overhead, use local_clock() instead. In order to test the difference between ktime_get() and local_clock(), a quick hack has been added to trigger, via debugfs, 1 times a call to ktime_get() and local_clock() and measure the elapsed time. Then the

[PATCH V2] cpuidle: Change ktime_get() with local_clock()

2016-04-21 Thread Daniel Lezcano
The ktime_get() can have a non negligeable overhead, use local_clock() instead. In order to test the difference between ktime_get() and local_clock(), a quick hack has been added to trigger, via debugfs, 1 times a call to ktime_get() and local_clock() and measure the elapsed time. Then the