Hi Yuyang,
On Mon, Jul 27, 2015 at 04:34:09AM +0800, Yuyang Du wrote:
> On Mon, Jul 27, 2015 at 12:04:20PM +0800, Boqun Feng wrote:
> > > ~~~
> > > > > 1) blocked load is more "difficult" to track, hint, migrate.
> > >
On Mon, Jul 27, 2015 at 12:04:20PM +0800, Boqun Feng wrote:
> > ~~~
> > > > 1) blocked load is more "difficult" to track, hint, migrate.
> > ~~~
>
> I may not get your point
Hi Yuyang,
On Mon, Jul 27, 2015 at 03:56:34AM +0800, Yuyang Du wrote:
> On Mon, Jul 27, 2015 at 11:21:15AM +0800, Boqun Feng wrote:
> > Hi Yuyang,
> >
> > On Mon, Jul 27, 2015 at 02:43:25AM +0800, Yuyang Du wrote:
> > > Hi Boqun,
> > >
> > > On Tue, Jul 21, 2015 at 06:29:56PM +0800, Boqun Feng
On Mon, Jul 27, 2015 at 11:21:15AM +0800, Boqun Feng wrote:
> Hi Yuyang,
>
> On Mon, Jul 27, 2015 at 02:43:25AM +0800, Yuyang Du wrote:
> > Hi Boqun,
> >
> > On Tue, Jul 21, 2015 at 06:29:56PM +0800, Boqun Feng wrote:
> > > The point is that you have already tracked the sum of runnable_load_avg
On Wed, Jul 22, 2015 at 10:19:54AM +0800, Boqun Feng wrote:
>
> > +/* Remove the runnable load generated by se from cfs_rq's runnable load
> > average */
> > +static inline void
> > +dequeue_entity_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se)
> > +{
> > + update_load_avg(se, 1);
>
On Mon, Jul 27, 2015 at 11:21:15AM +0800, Boqun Feng wrote:
> Hi Yuyang,
>
> On Mon, Jul 27, 2015 at 02:43:25AM +0800, Yuyang Du wrote:
> > Hi Boqun,
> >
> > On Tue, Jul 21, 2015 at 06:29:56PM +0800, Boqun Feng wrote:
> > > The point is that you have already tracked the sum of runnable_load_avg
Hi Yuyang,
On Mon, Jul 27, 2015 at 02:43:25AM +0800, Yuyang Du wrote:
> Hi Boqun,
>
> On Tue, Jul 21, 2015 at 06:29:56PM +0800, Boqun Feng wrote:
> > The point is that you have already tracked the sum of runnable_load_avg
> > and blocked_load_avg in cfs_rq->avg.load_avg. If you're going to track
Hi Boqun,
On Tue, Jul 21, 2015 at 06:29:56PM +0800, Boqun Feng wrote:
> The point is that you have already tracked the sum of runnable_load_avg
> and blocked_load_avg in cfs_rq->avg.load_avg. If you're going to track
> part of the sum, you'd better track the one that's updated less
> frequently,
Hi Yuyang,
On Mon, Jul 27, 2015 at 03:56:34AM +0800, Yuyang Du wrote:
On Mon, Jul 27, 2015 at 11:21:15AM +0800, Boqun Feng wrote:
Hi Yuyang,
On Mon, Jul 27, 2015 at 02:43:25AM +0800, Yuyang Du wrote:
Hi Boqun,
On Tue, Jul 21, 2015 at 06:29:56PM +0800, Boqun Feng wrote:
The
Hi Boqun,
On Tue, Jul 21, 2015 at 06:29:56PM +0800, Boqun Feng wrote:
The point is that you have already tracked the sum of runnable_load_avg
and blocked_load_avg in cfs_rq-avg.load_avg. If you're going to track
part of the sum, you'd better track the one that's updated less
frequently,
Hi Yuyang,
On Mon, Jul 27, 2015 at 02:43:25AM +0800, Yuyang Du wrote:
Hi Boqun,
On Tue, Jul 21, 2015 at 06:29:56PM +0800, Boqun Feng wrote:
The point is that you have already tracked the sum of runnable_load_avg
and blocked_load_avg in cfs_rq-avg.load_avg. If you're going to track
part
On Mon, Jul 27, 2015 at 11:21:15AM +0800, Boqun Feng wrote:
Hi Yuyang,
On Mon, Jul 27, 2015 at 02:43:25AM +0800, Yuyang Du wrote:
Hi Boqun,
On Tue, Jul 21, 2015 at 06:29:56PM +0800, Boqun Feng wrote:
The point is that you have already tracked the sum of runnable_load_avg
and
On Mon, Jul 27, 2015 at 11:21:15AM +0800, Boqun Feng wrote:
Hi Yuyang,
On Mon, Jul 27, 2015 at 02:43:25AM +0800, Yuyang Du wrote:
Hi Boqun,
On Tue, Jul 21, 2015 at 06:29:56PM +0800, Boqun Feng wrote:
The point is that you have already tracked the sum of runnable_load_avg
and
On Wed, Jul 22, 2015 at 10:19:54AM +0800, Boqun Feng wrote:
+/* Remove the runnable load generated by se from cfs_rq's runnable load
average */
+static inline void
+dequeue_entity_load_avg(struct cfs_rq *cfs_rq, struct sched_entity *se)
+{
+ update_load_avg(se, 1);
+
I think
On Mon, Jul 27, 2015 at 12:04:20PM +0800, Boqun Feng wrote:
~~~
1) blocked load is more difficult to track, hint, migrate.
~~~
I may not get your point here? Are you
Hi Yuyang,
On Mon, Jul 27, 2015 at 04:34:09AM +0800, Yuyang Du wrote:
On Mon, Jul 27, 2015 at 12:04:20PM +0800, Boqun Feng wrote:
~~~
1) blocked load is more difficult to track, hint, migrate.
On Wed, Jul 15, 2015 at 08:04:41AM +0800, Yuyang Du wrote:
> The cfs_rq's load_avg is composed of runnable_load_avg and blocked_load_avg.
> Before this series, sometimes the runnable_load_avg is used, and sometimes
> the load_avg is used. Completely replacing all uses of runnable_load_avg
> with
On Tue, Jul 21, 2015 at 06:18:46PM +0800, Boqun Feng wrote:
> On Tue, Jul 21, 2015 at 08:44:01AM +0800, Yuyang Du wrote:
> > On Tue, Jul 21, 2015 at 09:08:07AM +0800, Boqun Feng wrote:
> > > Hi Yuyang,
> > >
> > > On Wed, Jul 15, 2015 at 08:04:41AM +0800, Yuyang Du wrote:
> > > > The cfs_rq's
On Tue, Jul 21, 2015 at 08:44:01AM +0800, Yuyang Du wrote:
> On Tue, Jul 21, 2015 at 09:08:07AM +0800, Boqun Feng wrote:
> > Hi Yuyang,
> >
> > On Wed, Jul 15, 2015 at 08:04:41AM +0800, Yuyang Du wrote:
> > > The cfs_rq's load_avg is composed of runnable_load_avg and
> > > blocked_load_avg.
> >
On Tue, Jul 21, 2015 at 09:08:07AM +0800, Boqun Feng wrote:
> Hi Yuyang,
>
> On Wed, Jul 15, 2015 at 08:04:41AM +0800, Yuyang Du wrote:
> > The cfs_rq's load_avg is composed of runnable_load_avg and blocked_load_avg.
> > Before this series, sometimes the runnable_load_avg is used, and sometimes
>
On Wed, Jul 15, 2015 at 08:04:41AM +0800, Yuyang Du wrote:
The cfs_rq's load_avg is composed of runnable_load_avg and blocked_load_avg.
Before this series, sometimes the runnable_load_avg is used, and sometimes
the load_avg is used. Completely replacing all uses of runnable_load_avg
with
On Tue, Jul 21, 2015 at 09:08:07AM +0800, Boqun Feng wrote:
Hi Yuyang,
On Wed, Jul 15, 2015 at 08:04:41AM +0800, Yuyang Du wrote:
The cfs_rq's load_avg is composed of runnable_load_avg and blocked_load_avg.
Before this series, sometimes the runnable_load_avg is used, and sometimes
the
On Tue, Jul 21, 2015 at 08:44:01AM +0800, Yuyang Du wrote:
On Tue, Jul 21, 2015 at 09:08:07AM +0800, Boqun Feng wrote:
Hi Yuyang,
On Wed, Jul 15, 2015 at 08:04:41AM +0800, Yuyang Du wrote:
The cfs_rq's load_avg is composed of runnable_load_avg and
blocked_load_avg.
Before this
On Tue, Jul 21, 2015 at 06:18:46PM +0800, Boqun Feng wrote:
On Tue, Jul 21, 2015 at 08:44:01AM +0800, Yuyang Du wrote:
On Tue, Jul 21, 2015 at 09:08:07AM +0800, Boqun Feng wrote:
Hi Yuyang,
On Wed, Jul 15, 2015 at 08:04:41AM +0800, Yuyang Du wrote:
The cfs_rq's load_avg is
Hi Yuyang,
On Wed, Jul 15, 2015 at 08:04:41AM +0800, Yuyang Du wrote:
> The cfs_rq's load_avg is composed of runnable_load_avg and blocked_load_avg.
> Before this series, sometimes the runnable_load_avg is used, and sometimes
> the load_avg is used. Completely replacing all uses of
Hi Yuyang,
On Wed, Jul 15, 2015 at 08:04:41AM +0800, Yuyang Du wrote:
The cfs_rq's load_avg is composed of runnable_load_avg and blocked_load_avg.
Before this series, sometimes the runnable_load_avg is used, and sometimes
the load_avg is used. Completely replacing all uses of runnable_load_avg
The cfs_rq's load_avg is composed of runnable_load_avg and blocked_load_avg.
Before this series, sometimes the runnable_load_avg is used, and sometimes
the load_avg is used. Completely replacing all uses of runnable_load_avg
with load_avg may be too big a leap, i.e., the blocked_load_avg is
The cfs_rq's load_avg is composed of runnable_load_avg and blocked_load_avg.
Before this series, sometimes the runnable_load_avg is used, and sometimes
the load_avg is used. Completely replacing all uses of runnable_load_avg
with load_avg may be too big a leap, i.e., the blocked_load_avg is
28 matches
Mail list logo