On Wed, 2016-03-16 at 11:43 -0400, Chen, Tianyang wrote:
>
> On 03/16/2016 10:25 AM, Dario Faggioli wrote:
> >
> > This is too detailed for a changelog. If you want this information
> > to
> > live somewhere (it already lives in the list archives, actually),
> > make
> > a cover letter (this is
Ok, it's about time that we deal with this changelog!
On Mon, 2016-03-14 at 20:04 -0400, Tianyang Chen wrote:
> Budget replenishment and enforcement are separated by adding
> a replenishment timer, which fires at the next most imminent
> release time of all runnable vcpus.
>
First of all, state
On Wed, Mar 16, 2016 at 6:23 AM, Dario Faggioli
wrote:
> On Tue, 2016-03-15 at 23:40 -0400, Meng Xu wrote:
>> > > > @@ -115,6 +118,18 @@
>> > > > #define RTDS_delayed_runq_add (1<<__RTDS_delayed_runq_add)
>> > > >
>> > > > /*
>> > > > + * The replenishment timer
On 03/16/2016 10:25 AM, Dario Faggioli wrote:
Ok, it's about time that we deal with this changelog!
On Mon, 2016-03-14 at 20:04 -0400, Tianyang Chen wrote:
Budget replenishment and enforcement are separated by adding
a replenishment timer, which fires at the next most imminent
release time
On 03/16/2016 10:25 AM, Dario Faggioli wrote:
+if ( curr_on_cpu(vc->processor) == vc &&
>+ ( !list_empty(runq) ) )
>
So, this is because, since we don't keep the idle vcpus in the
runqueues, we need to catch the case where v is running, but no other
vcpu is waiting on the
On Wed, 2016-03-16 at 10:20 -0400, Meng Xu wrote:
> As to the comment, I will suggest:
>
> /*
> * RTDS_was_depleted: Is a vcpus budget depleted?
>
> * + Set in burn_budget() when a vcpus budget turns to zero
>
> * + Checked and cleared in repl_handler() to replenish the budget
>
> */
>
>
>> This is too detailed for a changelog. If you want this information to
>> live somewhere (it already lives in the list archives, actually), make
>> a cover letter (this is just one patch, so it's not required, but
>> nothing forbids that). Or put it in a wiki page. Or write a blog post.
>> Or
>>
>> sched_rt.c TIMER_SOFTIRQ
>> replenishment_timer_handler()
>> [spin_lock]
>> {
>> replenish(i)
>> runq_tickle(i)
>> }>
>> program_timer()
>> [spin_lock]
>>
> And kill this (or, again, move to cover,
On Wed, Mar 16, 2016 at 10:44 AM, Dario Faggioli
wrote:
> On Wed, 2016-03-16 at 10:20 -0400, Meng Xu wrote:
>> As to the comment, I will suggest:
>>
>> /*
>> * RTDS_was_depleted: Is a vcpus budget depleted?
>>
>> * + Set in burn_budget() when a vcpus budget turns to
On Tue, 2016-03-15 at 23:40 -0400, Meng Xu wrote:
> > > > @@ -115,6 +118,18 @@
> > > > #define RTDS_delayed_runq_add (1<<__RTDS_delayed_runq_add)
> > > >
> > > > /*
> > > > + * The replenishment timer handler needs to check this bit
> > > > + * to know where a replenished vcpu was, when
>>> @@ -115,6 +118,18 @@
>>> #define RTDS_delayed_runq_add (1<<__RTDS_delayed_runq_add)
>>>
>>> /*
>>> + * The replenishment timer handler needs to check this bit
>>> + * to know where a replenished vcpu was, when deciding which
>>> + * vcpu should tickle.
>>> + * A replenished vcpu should
On 03/15/2016 11:11 PM, Meng Xu wrote:
+
/*
* Flags
*/
@@ -115,6 +118,18 @@
#define RTDS_delayed_runq_add (1<<__RTDS_delayed_runq_add)
/*
+ * The replenishment timer handler needs to check this bit
+ * to know where a replenished vcpu was, when deciding which
+ * vcpu should
On Mon, Mar 14, 2016 at 8:04 PM, Tianyang Chen wrote:
> Budget replenishment and enforcement are separated by adding
> a replenishment timer, which fires at the next most imminent
> release time of all runnable vcpus.
>
> A replenishment queue has been added to keep track of
Budget replenishment and enforcement are separated by adding
a replenishment timer, which fires at the next most imminent
release time of all runnable vcpus.
A replenishment queue has been added to keep track of all vcpus that
are runnable.
The following functions have major changes to manage
14 matches
Mail list logo