Re: [PATCH RESEND v9 05/10] sched: make scale_rt invariant with frequency

2015-02-24 Thread Dietmar Eggemann
On 24/02/15 10:21, Vincent Guittot wrote: > On 19 February 2015 at 18:18, Morten Rasmussen > wrote: >> On Thu, Feb 19, 2015 at 04:52:41PM +, Peter Zijlstra wrote: >>> On Thu, Jan 15, 2015 at 11:09:25AM +0100, Vincent Guittot wrote: [...] >> Agreed. I think it is reasonable to assume that

Re: [PATCH RESEND v9 05/10] sched: make scale_rt invariant with frequency

2015-02-24 Thread Vincent Guittot
On 19 February 2015 at 18:18, Morten Rasmussen wrote: > On Thu, Feb 19, 2015 at 04:52:41PM +, Peter Zijlstra wrote: >> On Thu, Jan 15, 2015 at 11:09:25AM +0100, Vincent Guittot wrote: >> > The average running time of RT tasks is used to estimate the remaining >> > compute >> > capacity for

Re: [PATCH RESEND v9 05/10] sched: make scale_rt invariant with frequency

2015-02-24 Thread Vincent Guittot
On 19 February 2015 at 18:18, Morten Rasmussen morten.rasmus...@arm.com wrote: On Thu, Feb 19, 2015 at 04:52:41PM +, Peter Zijlstra wrote: On Thu, Jan 15, 2015 at 11:09:25AM +0100, Vincent Guittot wrote: The average running time of RT tasks is used to estimate the remaining compute

Re: [PATCH RESEND v9 05/10] sched: make scale_rt invariant with frequency

2015-02-24 Thread Dietmar Eggemann
On 24/02/15 10:21, Vincent Guittot wrote: On 19 February 2015 at 18:18, Morten Rasmussen morten.rasmus...@arm.com wrote: On Thu, Feb 19, 2015 at 04:52:41PM +, Peter Zijlstra wrote: On Thu, Jan 15, 2015 at 11:09:25AM +0100, Vincent Guittot wrote: [...] Agreed. I think it is reasonable

Re: [PATCH RESEND v9 05/10] sched: make scale_rt invariant with frequency

2015-02-19 Thread Morten Rasmussen
On Thu, Feb 19, 2015 at 04:52:41PM +, Peter Zijlstra wrote: > On Thu, Jan 15, 2015 at 11:09:25AM +0100, Vincent Guittot wrote: > > The average running time of RT tasks is used to estimate the remaining > > compute > > capacity for CFS tasks. This remaining capacity is the original capacity >

Re: [PATCH RESEND v9 05/10] sched: make scale_rt invariant with frequency

2015-02-19 Thread Peter Zijlstra
On Thu, Jan 15, 2015 at 11:09:25AM +0100, Vincent Guittot wrote: > The average running time of RT tasks is used to estimate the remaining compute > capacity for CFS tasks. This remaining capacity is the original capacity > scaled > down by a factor (aka scale_rt_capacity). This estimation of

Re: [PATCH RESEND v9 05/10] sched: make scale_rt invariant with frequency

2015-02-19 Thread Morten Rasmussen
On Thu, Feb 19, 2015 at 04:52:41PM +, Peter Zijlstra wrote: On Thu, Jan 15, 2015 at 11:09:25AM +0100, Vincent Guittot wrote: The average running time of RT tasks is used to estimate the remaining compute capacity for CFS tasks. This remaining capacity is the original capacity scaled

Re: [PATCH RESEND v9 05/10] sched: make scale_rt invariant with frequency

2015-02-19 Thread Peter Zijlstra
On Thu, Jan 15, 2015 at 11:09:25AM +0100, Vincent Guittot wrote: The average running time of RT tasks is used to estimate the remaining compute capacity for CFS tasks. This remaining capacity is the original capacity scaled down by a factor (aka scale_rt_capacity). This estimation of

[PATCH RESEND v9 05/10] sched: make scale_rt invariant with frequency

2015-01-15 Thread Vincent Guittot
The average running time of RT tasks is used to estimate the remaining compute capacity for CFS tasks. This remaining capacity is the original capacity scaled down by a factor (aka scale_rt_capacity). This estimation of available capacity must also be invariant with frequency scaling. A frequency

[PATCH RESEND v9 05/10] sched: make scale_rt invariant with frequency

2015-01-15 Thread Vincent Guittot
The average running time of RT tasks is used to estimate the remaining compute capacity for CFS tasks. This remaining capacity is the original capacity scaled down by a factor (aka scale_rt_capacity). This estimation of available capacity must also be invariant with frequency scaling. A frequency