Re: [PATCH 1/2 v2] sched: fix find_idlest_group for fork

2016-12-05 Thread Vincent Guittot
On 4 December 2016 at 00:25, Matt Fleming wrote: > On Fri, 25 Nov, at 04:34:32PM, Vincent Guittot wrote: >> During fork, the utilization of a task is init once the rq has been >> selected because the current utilization level of the rq is used to set >> the utilization

Re: [PATCH 1/2 v2] sched: fix find_idlest_group for fork

2016-12-05 Thread Vincent Guittot
On 4 December 2016 at 00:25, Matt Fleming wrote: > On Fri, 25 Nov, at 04:34:32PM, Vincent Guittot wrote: >> During fork, the utilization of a task is init once the rq has been >> selected because the current utilization level of the rq is used to set >> the utilization of the fork task. As the

Re: [PATCH 1/2 v2] sched: fix find_idlest_group for fork

2016-12-05 Thread Peter Zijlstra
On Tue, Nov 29, 2016 at 02:46:10PM +, Morten Rasmussen wrote: > On Tue, Nov 29, 2016 at 12:42:43PM +0100, Peter Zijlstra wrote: > > On Tue, Nov 29, 2016 at 10:57:59AM +, Morten Rasmussen wrote: > > > > @@ -5708,13 +5708,6 @@ static int select_idle_cpu(struct task_struct > > > > *p, struct

Re: [PATCH 1/2 v2] sched: fix find_idlest_group for fork

2016-12-05 Thread Peter Zijlstra
On Tue, Nov 29, 2016 at 02:46:10PM +, Morten Rasmussen wrote: > On Tue, Nov 29, 2016 at 12:42:43PM +0100, Peter Zijlstra wrote: > > On Tue, Nov 29, 2016 at 10:57:59AM +, Morten Rasmussen wrote: > > > > @@ -5708,13 +5708,6 @@ static int select_idle_cpu(struct task_struct > > > > *p, struct

Re: [PATCH 1/2 v2] sched: fix find_idlest_group for fork

2016-12-03 Thread Matt Fleming
On Fri, 25 Nov, at 04:34:32PM, Vincent Guittot wrote: > During fork, the utilization of a task is init once the rq has been > selected because the current utilization level of the rq is used to set > the utilization of the fork task. As the task's utilization is still > null at this step of the

Re: [PATCH 1/2 v2] sched: fix find_idlest_group for fork

2016-12-03 Thread Matt Fleming
On Fri, 25 Nov, at 04:34:32PM, Vincent Guittot wrote: > During fork, the utilization of a task is init once the rq has been > selected because the current utilization level of the rq is used to set > the utilization of the fork task. As the task's utilization is still > null at this step of the

Re: [PATCH 1/2 v2] sched: fix find_idlest_group for fork

2016-11-29 Thread Vincent Guittot
On 29 November 2016 at 15:50, Morten Rasmussen wrote: > On Tue, Nov 29, 2016 at 02:04:27PM +0100, Vincent Guittot wrote: >> On 29 November 2016 at 11:57, Morten Rasmussen >> wrote: >> > On Fri, Nov 25, 2016 at 04:34:32PM +0100, Vincent Guittot

Re: [PATCH 1/2 v2] sched: fix find_idlest_group for fork

2016-11-29 Thread Vincent Guittot
On 29 November 2016 at 15:50, Morten Rasmussen wrote: > On Tue, Nov 29, 2016 at 02:04:27PM +0100, Vincent Guittot wrote: >> On 29 November 2016 at 11:57, Morten Rasmussen >> wrote: >> > On Fri, Nov 25, 2016 at 04:34:32PM +0100, Vincent Guittot wrote: >> >> @@ -5708,13 +5708,6 @@ static int

Re: [PATCH 1/2 v2] sched: fix find_idlest_group for fork

2016-11-29 Thread Morten Rasmussen
On Tue, Nov 29, 2016 at 02:04:27PM +0100, Vincent Guittot wrote: > On 29 November 2016 at 11:57, Morten Rasmussen > wrote: > > On Fri, Nov 25, 2016 at 04:34:32PM +0100, Vincent Guittot wrote: > >> @@ -5708,13 +5708,6 @@ static int select_idle_cpu(struct task_struct *p,

Re: [PATCH 1/2 v2] sched: fix find_idlest_group for fork

2016-11-29 Thread Morten Rasmussen
On Tue, Nov 29, 2016 at 02:04:27PM +0100, Vincent Guittot wrote: > On 29 November 2016 at 11:57, Morten Rasmussen > wrote: > > On Fri, Nov 25, 2016 at 04:34:32PM +0100, Vincent Guittot wrote: > >> @@ -5708,13 +5708,6 @@ static int select_idle_cpu(struct task_struct *p, > >> struct sched_domain

Re: [PATCH 1/2 v2] sched: fix find_idlest_group for fork

2016-11-29 Thread Morten Rasmussen
On Tue, Nov 29, 2016 at 12:42:43PM +0100, Peter Zijlstra wrote: > On Tue, Nov 29, 2016 at 10:57:59AM +, Morten Rasmussen wrote: > > > @@ -5708,13 +5708,6 @@ static int select_idle_cpu(struct task_struct *p, > > > struct sched_domain *sd, int t > > > > > > avg_cost =

Re: [PATCH 1/2 v2] sched: fix find_idlest_group for fork

2016-11-29 Thread Morten Rasmussen
On Tue, Nov 29, 2016 at 12:42:43PM +0100, Peter Zijlstra wrote: > On Tue, Nov 29, 2016 at 10:57:59AM +, Morten Rasmussen wrote: > > > @@ -5708,13 +5708,6 @@ static int select_idle_cpu(struct task_struct *p, > > > struct sched_domain *sd, int t > > > > > > avg_cost =

Re: [PATCH 1/2 v2] sched: fix find_idlest_group for fork

2016-11-29 Thread Vincent Guittot
On 29 November 2016 at 11:57, Morten Rasmussen wrote: > On Fri, Nov 25, 2016 at 04:34:32PM +0100, Vincent Guittot wrote: >> During fork, the utilization of a task is init once the rq has been >> selected because the current utilization level of the rq is used to set >>

Re: [PATCH 1/2 v2] sched: fix find_idlest_group for fork

2016-11-29 Thread Vincent Guittot
On 29 November 2016 at 11:57, Morten Rasmussen wrote: > On Fri, Nov 25, 2016 at 04:34:32PM +0100, Vincent Guittot wrote: >> During fork, the utilization of a task is init once the rq has been >> selected because the current utilization level of the rq is used to set >> the utilization of the fork

Re: [PATCH 1/2 v2] sched: fix find_idlest_group for fork

2016-11-29 Thread Peter Zijlstra
On Tue, Nov 29, 2016 at 11:44:19AM +, Matt Fleming wrote: > On Tue, 29 Nov, at 12:42:43PM, Peter Zijlstra wrote: > > > > IIRC, and my pounding head really doesn't remember much, the comment > > reads like we need the large fudge factor because hackbench. That is, > > hackbench would like this

Re: [PATCH 1/2 v2] sched: fix find_idlest_group for fork

2016-11-29 Thread Peter Zijlstra
On Tue, Nov 29, 2016 at 11:44:19AM +, Matt Fleming wrote: > On Tue, 29 Nov, at 12:42:43PM, Peter Zijlstra wrote: > > > > IIRC, and my pounding head really doesn't remember much, the comment > > reads like we need the large fudge factor because hackbench. That is, > > hackbench would like this

Re: [PATCH 1/2 v2] sched: fix find_idlest_group for fork

2016-11-29 Thread Matt Fleming
On Tue, 29 Nov, at 12:42:43PM, Peter Zijlstra wrote: > > IIRC, and my pounding head really doesn't remember much, the comment > reads like we need the large fudge factor because hackbench. That is, > hackbench would like this test to go away, but others benchmarks will > tank. > > Now, if only I

Re: [PATCH 1/2 v2] sched: fix find_idlest_group for fork

2016-11-29 Thread Matt Fleming
On Tue, 29 Nov, at 12:42:43PM, Peter Zijlstra wrote: > > IIRC, and my pounding head really doesn't remember much, the comment > reads like we need the large fudge factor because hackbench. That is, > hackbench would like this test to go away, but others benchmarks will > tank. > > Now, if only I

Re: [PATCH 1/2 v2] sched: fix find_idlest_group for fork

2016-11-29 Thread Peter Zijlstra
On Tue, Nov 29, 2016 at 10:57:59AM +, Morten Rasmussen wrote: > > @@ -5708,13 +5708,6 @@ static int select_idle_cpu(struct task_struct *p, > > struct sched_domain *sd, int t > > > > avg_cost = this_sd->avg_scan_cost; > > > > - /* > > -* Due to large variance we need a large fuzz

Re: [PATCH 1/2 v2] sched: fix find_idlest_group for fork

2016-11-29 Thread Peter Zijlstra
On Tue, Nov 29, 2016 at 10:57:59AM +, Morten Rasmussen wrote: > > @@ -5708,13 +5708,6 @@ static int select_idle_cpu(struct task_struct *p, > > struct sched_domain *sd, int t > > > > avg_cost = this_sd->avg_scan_cost; > > > > - /* > > -* Due to large variance we need a large fuzz

Re: [PATCH 1/2 v2] sched: fix find_idlest_group for fork

2016-11-29 Thread Morten Rasmussen
On Fri, Nov 25, 2016 at 04:34:32PM +0100, Vincent Guittot wrote: > During fork, the utilization of a task is init once the rq has been > selected because the current utilization level of the rq is used to set > the utilization of the fork task. As the task's utilization is still > null at this

Re: [PATCH 1/2 v2] sched: fix find_idlest_group for fork

2016-11-29 Thread Morten Rasmussen
On Fri, Nov 25, 2016 at 04:34:32PM +0100, Vincent Guittot wrote: > During fork, the utilization of a task is init once the rq has been > selected because the current utilization level of the rq is used to set > the utilization of the fork task. As the task's utilization is still > null at this

Re: [PATCH 1/2 v2] sched: fix find_idlest_group for fork

2016-11-28 Thread Vincent Guittot
On 28 November 2016 at 18:01, Matt Fleming wrote: > On Fri, 25 Nov, at 04:34:32PM, Vincent Guittot wrote: >> >> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c >> index aa47589..820a787 100644 >> --- a/kernel/sched/fair.c >> +++ b/kernel/sched/fair.c >> @@

Re: [PATCH 1/2 v2] sched: fix find_idlest_group for fork

2016-11-28 Thread Vincent Guittot
On 28 November 2016 at 18:01, Matt Fleming wrote: > On Fri, 25 Nov, at 04:34:32PM, Vincent Guittot wrote: >> >> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c >> index aa47589..820a787 100644 >> --- a/kernel/sched/fair.c >> +++ b/kernel/sched/fair.c >> @@ -5463,13 +5463,19 @@

Re: [PATCH 1/2 v2] sched: fix find_idlest_group for fork

2016-11-28 Thread Matt Fleming
On Fri, 25 Nov, at 04:34:32PM, Vincent Guittot wrote: > > diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c > index aa47589..820a787 100644 > --- a/kernel/sched/fair.c > +++ b/kernel/sched/fair.c > @@ -5463,13 +5463,19 @@ find_idlest_group(struct sched_domain *sd, struct > task_struct *p, >

Re: [PATCH 1/2 v2] sched: fix find_idlest_group for fork

2016-11-28 Thread Matt Fleming
On Fri, 25 Nov, at 04:34:32PM, Vincent Guittot wrote: > > diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c > index aa47589..820a787 100644 > --- a/kernel/sched/fair.c > +++ b/kernel/sched/fair.c > @@ -5463,13 +5463,19 @@ find_idlest_group(struct sched_domain *sd, struct > task_struct *p, >