On Thu, Jan 24, 2019 at 11:21:53AM +, Patrick Bellasi wrote:
> When a task-specific uclamp value is changed for a task, instead, a
> dequeue/enqueue is not needed. As long as we are doing a lazy update,
> that sounds just like not necessary overhead.
When that overhead is shown to be a proble
On 23-Jan 19:59, Peter Zijlstra wrote:
> On Wed, Jan 23, 2019 at 02:14:26PM +, Patrick Bellasi wrote:
>
> > > > Consider also that the uclamp_task_update_active() added by this patch
> > > > not only has lower overhead but it will be use also by cgroups where
> > > > we want to force update al
On Wed, Jan 23, 2019 at 02:14:26PM +, Patrick Bellasi wrote:
> > > Consider also that the uclamp_task_update_active() added by this patch
> > > not only has lower overhead but it will be use also by cgroups where
> > > we want to force update all the tasks on a cgroup's clamp change.
> >
> >
On 23-Jan 10:16, Peter Zijlstra wrote:
> On Tue, Jan 22, 2019 at 03:33:15PM +, Patrick Bellasi wrote:
> > On 22-Jan 15:57, Peter Zijlstra wrote:
> > > On Tue, Jan 22, 2019 at 02:01:15PM +, Patrick Bellasi wrote:
>
> > > > Yes, I would say we have two options:
> > > >
> > > > 1) SCHED_FLA
On Tue, Jan 22, 2019 at 03:33:15PM +, Patrick Bellasi wrote:
> On 22-Jan 15:57, Peter Zijlstra wrote:
> > On Tue, Jan 22, 2019 at 02:01:15PM +, Patrick Bellasi wrote:
> > > Yes, I would say we have two options:
> > >
> > > 1) SCHED_FLAG_KEEP_POLICY enforces all the scheduling class speci
On 22-Jan 15:57, Peter Zijlstra wrote:
> On Tue, Jan 22, 2019 at 02:01:15PM +, Patrick Bellasi wrote:
> > On 22-Jan 14:28, Peter Zijlstra wrote:
> > > On Tue, Jan 22, 2019 at 10:43:05AM +, Patrick Bellasi wrote:
> > > > On 22-Jan 10:37, Peter Zijlstra wrote:
> > >
> > > > > Sure, I get tha
On Tue, Jan 22, 2019 at 02:01:15PM +, Patrick Bellasi wrote:
> On 22-Jan 14:28, Peter Zijlstra wrote:
> > On Tue, Jan 22, 2019 at 10:43:05AM +, Patrick Bellasi wrote:
> > > On 22-Jan 10:37, Peter Zijlstra wrote:
> >
> > > > Sure, I get that. What I don't get is why you're adding that (2) h
On 22-Jan 14:28, Peter Zijlstra wrote:
> On Tue, Jan 22, 2019 at 10:43:05AM +, Patrick Bellasi wrote:
> > On 22-Jan 10:37, Peter Zijlstra wrote:
>
> > > Sure, I get that. What I don't get is why you're adding that (2) here.
> > > Like said, __sched_setscheduler() already does a dequeue/enqueue
On Tue, Jan 22, 2019 at 10:43:05AM +, Patrick Bellasi wrote:
> On 22-Jan 10:37, Peter Zijlstra wrote:
> > Sure, I get that. What I don't get is why you're adding that (2) here.
> > Like said, __sched_setscheduler() already does a dequeue/enqueue under
> > rq->lock, which should already take ca
On 22-Jan 10:37, Peter Zijlstra wrote:
> On Mon, Jan 21, 2019 at 03:44:12PM +, Patrick Bellasi wrote:
> > On 21-Jan 16:33, Peter Zijlstra wrote:
> > > On Tue, Jan 15, 2019 at 10:15:02AM +, Patrick Bellasi wrote:
> > >
> > > > +static inline void
> > > > +uclamp_task_update_active(struct ta
On Mon, Jan 21, 2019 at 03:44:12PM +, Patrick Bellasi wrote:
> On 21-Jan 16:33, Peter Zijlstra wrote:
> > On Tue, Jan 15, 2019 at 10:15:02AM +, Patrick Bellasi wrote:
> >
> > > +static inline void
> > > +uclamp_task_update_active(struct task_struct *p, unsigned int clamp_id)
> > > +{
> > >
On 21-Jan 16:33, Peter Zijlstra wrote:
> On Tue, Jan 15, 2019 at 10:15:02AM +, Patrick Bellasi wrote:
>
> > +static inline void
> > +uclamp_task_update_active(struct task_struct *p, unsigned int clamp_id)
> > +{
> > + struct rq_flags rf;
> > + struct rq *rq;
> > +
> > + /*
> > +* Loc
On Tue, Jan 15, 2019 at 10:15:02AM +, Patrick Bellasi wrote:
> +static inline void
> +uclamp_task_update_active(struct task_struct *p, unsigned int clamp_id)
> +{
> + struct rq_flags rf;
> + struct rq *rq;
> +
> + /*
> + * Lock the task and the CPU where the task is (or was) q
Utilization clamp values enforced on a CPU by a task can be updated, for
example via a sched_setattr() syscall, while a task is RUNNABLE on that
CPU. A clamp value change always implies a clamp bucket refcount update
to ensure the new constraints are enforced.
Hook into uclamp_bucket_get() to trig
14 matches
Mail list logo
