Re: [LKP] [lkp-developer] [sched/fair] 4e5160766f: +149% ftq.noise.50% regression

2017-02-28 Thread Vincent Guittot
On 28 February 2017 at 01:33, Huang, Ying wrote: > Vincent Guittot writes: > >> Hi Ying, >> >> On 21 February 2017 at 03:40, Huang, Ying wrote: >>> Hi, Vincent, >>> >>> Vincent Guittot writes: >>> >> >> [snip] >> > > Here is the test result, > > =

Re: [LKP] [lkp-developer] [sched/fair] 4e5160766f: +149% ftq.noise.50% regression

2017-02-27 Thread Huang, Ying
Vincent Guittot writes: > Hi Ying, > > On 21 February 2017 at 03:40, Huang, Ying wrote: >> Hi, Vincent, >> >> Vincent Guittot writes: >> > > [snip] > Here is the test result, = compi

Re: [LKP] [lkp-developer] [sched/fair] 4e5160766f: +149% ftq.noise.50% regression

2017-02-27 Thread Vincent Guittot
Hi Ying, On 21 February 2017 at 03:40, Huang, Ying wrote: > Hi, Vincent, > > Vincent Guittot writes: > [snip] >>> >>> Here is the test result, >>> >>> = >>> compiler/cpufreq_governor/freq/kconfig/nr_task/ro

Re: [LKP] [lkp-developer] [sched/fair] 4e5160766f: +149% ftq.noise.50% regression

2017-02-20 Thread Huang, Ying
Hi, Vincent, Vincent Guittot writes: > On 4 January 2017 at 04:08, Huang, Ying wrote: >> Vincent Guittot writes: >> Vincent, like we discussed in September last year, the proper fix would probably be a cfs-rq->nr_attached which IMHO is not doable w/o being an atomic because

Re: [LKP] [lkp-developer] [sched/fair] 4e5160766f: +149% ftq.noise.50% regression

2017-01-04 Thread Vincent Guittot
On 4 January 2017 at 04:08, Huang, Ying wrote: > Vincent Guittot writes: > >>> >>> Vincent, like we discussed in September last year, the proper fix would >>> probably be a cfs-rq->nr_attached which IMHO is not doable w/o being an >>> atomic because of migrate_task_rq_fair()->remove_entity_load_a

Re: [LKP] [lkp-developer] [sched/fair] 4e5160766f: +149% ftq.noise.50% regression

2017-01-03 Thread Huang, Ying
Vincent Guittot writes: > Hi Dietmar and Ying, > > Le Tuesday 03 Jan 2017 11:38:39 (+0100), Dietmar Eggemann a crit : >> Hi Vincent and Ying, >> >> On 01/02/2017 04:42 PM, Vincent Guittot wrote: >> >Hi Ying, >> > >> >On 28 December 2016 at 09:17, Huang, Ying wrote: >> >>Vincent Guittot writes

Re: [LKP] [lkp-developer] [sched/fair] 4e5160766f: +149% ftq.noise.50% regression

2017-01-03 Thread Vincent Guittot
Hi Dietmar and Ying, Le Tuesday 03 Jan 2017 à 11:38:39 (+0100), Dietmar Eggemann a écrit : > Hi Vincent and Ying, > > On 01/02/2017 04:42 PM, Vincent Guittot wrote: > >Hi Ying, > > > >On 28 December 2016 at 09:17, Huang, Ying wrote: > >>Vincent Guittot writes: > >> > >>>Le Tuesday 13 Dec 2016 .

Re: [LKP] [lkp-developer] [sched/fair] 4e5160766f: +149% ftq.noise.50% regression

2017-01-03 Thread Dietmar Eggemann
Hi Vincent and Ying, On 01/02/2017 04:42 PM, Vincent Guittot wrote: Hi Ying, On 28 December 2016 at 09:17, Huang, Ying wrote: Vincent Guittot writes: Le Tuesday 13 Dec 2016 . 09:47:30 (+0800), Huang, Ying a .crit : Hi, Vincent, Vincent Guittot writes: [...] --- kernel/sched/fair.c

Re: [LKP] [lkp-developer] [sched/fair] 4e5160766f: +149% ftq.noise.50% regression

2017-01-02 Thread Vincent Guittot
Hi Ying, On 28 December 2016 at 09:17, Huang, Ying wrote: > Vincent Guittot writes: > >> Le Tuesday 13 Dec 2016 . 09:47:30 (+0800), Huang, Ying a .crit : >>> Hi, Vincent, >>> >>> Vincent Guittot writes: >>> >>> > Hi Ying, >>> > >>> > On 12 December 2016 at 06:43, kernel test robot >>> > wrote:

Re: [LKP] [lkp-developer] [sched/fair] 4e5160766f: +149% ftq.noise.50% regression

2016-12-28 Thread Huang, Ying
Vincent Guittot writes: > Le Tuesday 13 Dec 2016 . 09:47:30 (+0800), Huang, Ying a .crit : >> Hi, Vincent, >> >> Vincent Guittot writes: >> >> > Hi Ying, >> > >> > On 12 December 2016 at 06:43, kernel test robot >> > wrote: >> >> Greeting, >> >> >> >> FYI, we noticed a 149% regression of ftq.

Re: [LKP] [lkp-developer] [sched/fair] 4e5160766f: +149% ftq.noise.50% regression

2016-12-22 Thread Vincent Guittot
Le Tuesday 13 Dec 2016 à 09:47:30 (+0800), Huang, Ying a écrit : > Hi, Vincent, > > Vincent Guittot writes: > > > Hi Ying, > > > > On 12 December 2016 at 06:43, kernel test robot > > wrote: > >> Greeting, > >> > >> FYI, we noticed a 149% regression of ftq.noise.50% due to commit: > >> > >> > >>

Re: [LKP] [lkp-developer] [sched/fair] 4e5160766f: +149% ftq.noise.50% regression

2016-12-12 Thread Huang, Ying
Hi, Vincent, Vincent Guittot writes: > Hi Ying, > > On 12 December 2016 at 06:43, kernel test robot > wrote: >> Greeting, >> >> FYI, we noticed a 149% regression of ftq.noise.50% due to commit: >> >> >> commit: 4e5160766fcc9f41bbd38bac11f92dce993644aa ("sched/fair: Propagate >> asynchrous deta