Re: [RFC PATCH 1/2] sched/rt: add utilization tracking

On Wed, May 31, 2017 at 11:41:11AM +0100, Juri Lelli wrote: > On 31/05/17 12:30, Peter Zijlstra wrote: > > Again, it seems I have this CPPC/HWP crud firmly stuck in my brain. > > Because I was thinking: > > > > min_freq = dl_util > > avg_freq = dl_avg + rt_avg + cfs_util > > > > > >

Re: [RFC PATCH 1/2] sched/rt: add utilization tracking

On Wed, May 31, 2017 at 11:41:11AM +0100, Juri Lelli wrote: > On 31/05/17 12:30, Peter Zijlstra wrote: > > Again, it seems I have this CPPC/HWP crud firmly stuck in my brain. > > Because I was thinking: > > > > min_freq = dl_util > > avg_freq = dl_avg + rt_avg + cfs_util > > > > > >

Re: [RFC PATCH 1/2] sched/rt: add utilization tracking

On 31 May 2017 at 11:40, Peter Zijlstra wrote: > On Wed, May 24, 2017 at 11:00:51AM +0200, Vincent Guittot wrote: >> schedutil governor relies on cfs_rq's util_avg to choose the OPP when cfs >> tasks are running. When the CPU is overloaded by cfs and rt tasks, cfs tasks >>

Re: [RFC PATCH 1/2] sched/rt: add utilization tracking

On 31 May 2017 at 11:40, Peter Zijlstra wrote: > On Wed, May 24, 2017 at 11:00:51AM +0200, Vincent Guittot wrote: >> schedutil governor relies on cfs_rq's util_avg to choose the OPP when cfs >> tasks are running. When the CPU is overloaded by cfs and rt tasks, cfs tasks >> are preempted by rt

Re: [RFC PATCH 1/2] sched/rt: add utilization tracking

On 31/05/17 12:30, Peter Zijlstra wrote: > On Wed, May 31, 2017 at 11:40:47AM +0200, Peter Zijlstra wrote: > > On Wed, May 24, 2017 at 11:00:51AM +0200, Vincent Guittot wrote: > > > schedutil governor relies on cfs_rq's util_avg to choose the OPP when cfs > > > tasks are running. When the CPU is

Re: [RFC PATCH 1/2] sched/rt: add utilization tracking

On 31/05/17 12:30, Peter Zijlstra wrote: > On Wed, May 31, 2017 at 11:40:47AM +0200, Peter Zijlstra wrote: > > On Wed, May 24, 2017 at 11:00:51AM +0200, Vincent Guittot wrote: > > > schedutil governor relies on cfs_rq's util_avg to choose the OPP when cfs > > > tasks are running. When the CPU is

Re: [RFC PATCH 1/2] sched/rt: add utilization tracking

On Wed, May 31, 2017 at 11:40:47AM +0200, Peter Zijlstra wrote: > On Wed, May 24, 2017 at 11:00:51AM +0200, Vincent Guittot wrote: > > schedutil governor relies on cfs_rq's util_avg to choose the OPP when cfs > > tasks are running. When the CPU is overloaded by cfs and rt tasks, cfs tasks > > are

Re: [RFC PATCH 1/2] sched/rt: add utilization tracking

On Wed, May 31, 2017 at 11:40:47AM +0200, Peter Zijlstra wrote: > On Wed, May 24, 2017 at 11:00:51AM +0200, Vincent Guittot wrote: > > schedutil governor relies on cfs_rq's util_avg to choose the OPP when cfs > > tasks are running. When the CPU is overloaded by cfs and rt tasks, cfs tasks > > are

Re: [RFC PATCH 1/2] sched/rt: add utilization tracking

On Wed, May 24, 2017 at 11:00:51AM +0200, Vincent Guittot wrote: > schedutil governor relies on cfs_rq's util_avg to choose the OPP when cfs > tasks are running. When the CPU is overloaded by cfs and rt tasks, cfs tasks > are preempted by rt tasks and in this case util_avg reflects the remaining >

Re: [RFC PATCH 1/2] sched/rt: add utilization tracking

On Wed, May 24, 2017 at 11:00:51AM +0200, Vincent Guittot wrote: > schedutil governor relies on cfs_rq's util_avg to choose the OPP when cfs > tasks are running. When the CPU is overloaded by cfs and rt tasks, cfs tasks > are preempted by rt tasks and in this case util_avg reflects the remaining >

Re: [RFC PATCH 1/2] sched/rt: add utilization tracking

On Tue, May 30, 2017 at 06:24:27PM +0200, Vincent Guittot wrote: > On 30 May 2017 at 17:50, Morten Rasmussen wrote: > > On Wed, May 24, 2017 at 11:00:51AM +0200, Vincent Guittot wrote: > >> @@ -1534,6 +1534,8 @@ static struct task_struct *_pick_next_task_rt(struct > >>

Re: [RFC PATCH 1/2] sched/rt: add utilization tracking

On Tue, May 30, 2017 at 06:24:27PM +0200, Vincent Guittot wrote: > On 30 May 2017 at 17:50, Morten Rasmussen wrote: > > On Wed, May 24, 2017 at 11:00:51AM +0200, Vincent Guittot wrote: > >> @@ -1534,6 +1534,8 @@ static struct task_struct *_pick_next_task_rt(struct > >> rq *rq) > >> return

Re: [RFC PATCH 1/2] sched/rt: add utilization tracking

On 30 May 2017 at 17:50, Morten Rasmussen wrote: > On Wed, May 24, 2017 at 11:00:51AM +0200, Vincent Guittot wrote: >> schedutil governor relies on cfs_rq's util_avg to choose the OPP when cfs >> tasks are running. When the CPU is overloaded by cfs and rt tasks, cfs

Re: [RFC PATCH 1/2] sched/rt: add utilization tracking

On 30 May 2017 at 17:50, Morten Rasmussen wrote: > On Wed, May 24, 2017 at 11:00:51AM +0200, Vincent Guittot wrote: >> schedutil governor relies on cfs_rq's util_avg to choose the OPP when cfs >> tasks are running. When the CPU is overloaded by cfs and rt tasks, cfs tasks >> are preempted by rt

Re: [RFC PATCH 1/2] sched/rt: add utilization tracking

On Wed, May 24, 2017 at 11:00:51AM +0200, Vincent Guittot wrote: > schedutil governor relies on cfs_rq's util_avg to choose the OPP when cfs > tasks are running. When the CPU is overloaded by cfs and rt tasks, cfs tasks > are preempted by rt tasks and in this case util_avg reflects the remaining >

Re: [RFC PATCH 1/2] sched/rt: add utilization tracking

On Wed, May 24, 2017 at 11:00:51AM +0200, Vincent Guittot wrote: > schedutil governor relies on cfs_rq's util_avg to choose the OPP when cfs > tasks are running. When the CPU is overloaded by cfs and rt tasks, cfs tasks > are preempted by rt tasks and in this case util_avg reflects the remaining >

[RFC PATCH 1/2] sched/rt: add utilization tracking

schedutil governor relies on cfs_rq's util_avg to choose the OPP when cfs tasks are running. When the CPU is overloaded by cfs and rt tasks, cfs tasks are preempted by rt tasks and in this case util_avg reflects the remaining capacity that is used by cfs tasks but not what cfs tasks want to use.

[RFC PATCH 1/2] sched/rt: add utilization tracking

schedutil governor relies on cfs_rq's util_avg to choose the OPP when cfs tasks are running. When the CPU is overloaded by cfs and rt tasks, cfs tasks are preempted by rt tasks and in this case util_avg reflects the remaining capacity that is used by cfs tasks but not what cfs tasks want to use.