Re: [PATCH] sched/cpufreq: don't trigger cpufreq update w/o real rt/deadline tasks running

2016-04-21 Thread Peter Zijlstra
On Thu, Apr 21, 2016 at 07:07:51PM +0200, Rafael J. Wysocki wrote: > On Thu, Apr 21, 2016 at 2:33 PM, Peter Zijlstra wrote: > > On Thu, Apr 21, 2016 at 09:09:43AM +0800, Wanpeng Li wrote: > >> >> Sometimes .update_curr hook is called w/o tasks actually running, it is > >> >> captured by: > >> >> >

Re: [PATCH] sched/cpufreq: don't trigger cpufreq update w/o real rt/deadline tasks running

2016-04-21 Thread Rafael J. Wysocki
On Thu, Apr 21, 2016 at 2:33 PM, Peter Zijlstra wrote: > On Thu, Apr 21, 2016 at 09:09:43AM +0800, Wanpeng Li wrote: >> >> Sometimes .update_curr hook is called w/o tasks actually running, it is >> >> captured by: >> >> >> >> u64 delta_exec = rq_clock_task(rq) - curr->se.exec_start; >> >>

Re: [PATCH] sched/cpufreq: don't trigger cpufreq update w/o real rt/deadline tasks running

2016-04-21 Thread Wanpeng Li
Hi Peterz, 2016-04-21 20:33 GMT+08:00 Peter Zijlstra : > On Thu, Apr 21, 2016 at 09:09:43AM +0800, Wanpeng Li wrote: >> >> Sometimes .update_curr hook is called w/o tasks actually running, it is >> >> captured by: >> >> >> >> u64 delta_exec = rq_clock_task(rq) - curr->se.exec_start; >> >>

Re: [PATCH] sched/cpufreq: don't trigger cpufreq update w/o real rt/deadline tasks running

2016-04-21 Thread Peter Zijlstra
On Thu, Apr 21, 2016 at 09:09:43AM +0800, Wanpeng Li wrote: > >> Sometimes .update_curr hook is called w/o tasks actually running, it is > >> captured by: > >> > >> u64 delta_exec = rq_clock_task(rq) - curr->se.exec_start; > >> > >> We should not trigger cpufreq update in this case for rt/

Re: [PATCH] sched/cpufreq: don't trigger cpufreq update w/o real rt/deadline tasks running

2016-04-21 Thread Wanpeng Li
2016-04-21 20:12 GMT+08:00 Wanpeng Li : > 2016-04-21 19:11 GMT+08:00 Rafael J. Wysocki : >> On 4/21/2016 3:09 AM, Wanpeng Li wrote: >>> >>> 2016-04-21 6:28 GMT+08:00 Rafael J. Wysocki : On 4/21/2016 12:24 AM, Wanpeng Li wrote: > > 2016-04-20 22:01 GMT+08:00 Peter Zijlstra : >>

Re: [PATCH] sched/cpufreq: don't trigger cpufreq update w/o real rt/deadline tasks running

2016-04-21 Thread Wanpeng Li
2016-04-21 19:11 GMT+08:00 Rafael J. Wysocki : > On 4/21/2016 3:09 AM, Wanpeng Li wrote: >> >> 2016-04-21 6:28 GMT+08:00 Rafael J. Wysocki : >>> >>> On 4/21/2016 12:24 AM, Wanpeng Li wrote: 2016-04-20 22:01 GMT+08:00 Peter Zijlstra : > > On Wed, Apr 20, 2016 at 02:32:35AM +0200, R

Re: [PATCH] sched/cpufreq: don't trigger cpufreq update w/o real rt/deadline tasks running

2016-04-21 Thread Rafael J. Wysocki
On 4/21/2016 3:09 AM, Wanpeng Li wrote: 2016-04-21 6:28 GMT+08:00 Rafael J. Wysocki : On 4/21/2016 12:24 AM, Wanpeng Li wrote: 2016-04-20 22:01 GMT+08:00 Peter Zijlstra : On Wed, Apr 20, 2016 at 02:32:35AM +0200, Rafael J. Wysocki wrote: On Monday, April 18, 2016 01:51:24 PM Wanpeng Li wrote:

Re: [PATCH] sched/cpufreq: don't trigger cpufreq update w/o real rt/deadline tasks running

2016-04-20 Thread Wanpeng Li
2016-04-21 6:28 GMT+08:00 Rafael J. Wysocki : > On 4/21/2016 12:24 AM, Wanpeng Li wrote: >> >> 2016-04-20 22:01 GMT+08:00 Peter Zijlstra : >>> >>> On Wed, Apr 20, 2016 at 02:32:35AM +0200, Rafael J. Wysocki wrote: On Monday, April 18, 2016 01:51:24 PM Wanpeng Li wrote: > > Sometim

Re: [PATCH] sched/cpufreq: don't trigger cpufreq update w/o real rt/deadline tasks running

2016-04-20 Thread Rafael J. Wysocki
On 4/21/2016 12:24 AM, Wanpeng Li wrote: 2016-04-20 22:01 GMT+08:00 Peter Zijlstra : On Wed, Apr 20, 2016 at 02:32:35AM +0200, Rafael J. Wysocki wrote: On Monday, April 18, 2016 01:51:24 PM Wanpeng Li wrote: Sometimes update_curr() is called w/o tasks actually running, it is captured by:

Re: [PATCH] sched/cpufreq: don't trigger cpufreq update w/o real rt/deadline tasks running

2016-04-20 Thread Wanpeng Li
2016-04-20 22:01 GMT+08:00 Peter Zijlstra : > On Wed, Apr 20, 2016 at 02:32:35AM +0200, Rafael J. Wysocki wrote: >> On Monday, April 18, 2016 01:51:24 PM Wanpeng Li wrote: >> > Sometimes update_curr() is called w/o tasks actually running, it is >> > captured by: >> > u64 delta_exec = rq_clock_t

Re: [PATCH] sched/cpufreq: don't trigger cpufreq update w/o real rt/deadline tasks running

2016-04-20 Thread Peter Zijlstra
On Wed, Apr 20, 2016 at 02:32:35AM +0200, Rafael J. Wysocki wrote: > On Monday, April 18, 2016 01:51:24 PM Wanpeng Li wrote: > > Sometimes update_curr() is called w/o tasks actually running, it is > > captured by: > > u64 delta_exec = rq_clock_task(rq) - curr->se.exec_start; > > We should not

Re: [PATCH] sched/cpufreq: don't trigger cpufreq update w/o real rt/deadline tasks running

2016-04-19 Thread Wanpeng Li
2016-04-20 8:32 GMT+08:00 Rafael J. Wysocki : > On Monday, April 18, 2016 01:51:24 PM Wanpeng Li wrote: >> Sometimes update_curr() is called w/o tasks actually running, it is >> captured by: >> u64 delta_exec = rq_clock_task(rq) - curr->se.exec_start; >> We should not trigger cpufreq update i

Re: [PATCH] sched/cpufreq: don't trigger cpufreq update w/o real rt/deadline tasks running

2016-04-19 Thread Rafael J. Wysocki
On Monday, April 18, 2016 01:51:24 PM Wanpeng Li wrote: > Sometimes update_curr() is called w/o tasks actually running, it is > captured by: > u64 delta_exec = rq_clock_task(rq) - curr->se.exec_start; > We should not trigger cpufreq update in this case for rt/deadline > classes, and this pat

[PATCH] sched/cpufreq: don't trigger cpufreq update w/o real rt/deadline tasks running

2016-04-17 Thread Wanpeng Li
Sometimes update_curr() is called w/o tasks actually running, it is captured by: u64 delta_exec = rq_clock_task(rq) - curr->se.exec_start; We should not trigger cpufreq update in this case for rt/deadline classes, and this patch fix it. Signed-off-by: Wanpeng Li --- kernel/sched/deadlin