On Thu, May 06, 2021 at 07:09:36PM +0800, Xuewen Yan wrote:
> From: Xuewen Yan <[email protected]>
>
> The UTIL_AVG_UNCHANGED flag had been cleared when the task util changed.
> And the enqueued is equal to task_util with the flag, so it is better
> to add the UTIL_AVG_UNCHANGED flag for last_enqueued_diff.
>
> Fixes: b89997aa88f0b sched/pelt: Fix task util_est update filtering
>
> Signed-off-by: Xuewen Yan <[email protected]>
> ---
> kernel/sched/fair.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index e5e457fa9dc8..94d77b4fa601 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -3996,7 +3996,7 @@ static inline void util_est_update(struct cfs_rq *cfs_rq,
> if (ue.enqueued & UTIL_AVG_UNCHANGED)
> return;
>
> - last_enqueued_diff = ue.enqueued;
> + last_enqueued_diff = (ue.enqueued | UTIL_AVG_UNCHANGED);
>
> /*
> * Reset EWMA on utilization increases, the moving average is used only
> --
> 2.29.0
>
Hi,
We do indeed for the diff use the flag for the value updated and no flag for the
value before the update. However, last_enqueued_diff is only used for the margin
check which is an heuristic and is not an accurate value (~1%) and as we know
we already loose the LSB in util_est, I'm not sure this is really necessary.
--
Vincent
Hi
On Thu, May 6, 2021 at 8:28 PM Vincent Donnefort
<[email protected]> wrote:
>
> On Thu, May 06, 2021 at 07:09:36PM +0800, Xuewen Yan wrote:
> > From: Xuewen Yan <[email protected]>
> >
> > The UTIL_AVG_UNCHANGED flag had been cleared when the task util changed.
> > And the enqueued is equal to task_util with the flag, so it is better
> > to add the UTIL_AVG_UNCHANGED flag for last_enqueued_diff.
> >
> > Fixes: b89997aa88f0b sched/pelt: Fix task util_est update filtering
> >
> > Signed-off-by: Xuewen Yan <[email protected]>
> > ---
> > kernel/sched/fair.c | 2 +-
> > 1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> > index e5e457fa9dc8..94d77b4fa601 100644
> > --- a/kernel/sched/fair.c
> > +++ b/kernel/sched/fair.c
> > @@ -3996,7 +3996,7 @@ static inline void util_est_update(struct cfs_rq *cfs_rq,
> > if (ue.enqueued & UTIL_AVG_UNCHANGED)
> > return;
> >
> > - last_enqueued_diff = ue.enqueued;
> > + last_enqueued_diff = (ue.enqueued | UTIL_AVG_UNCHANGED);
> >
> > /*
> > * Reset EWMA on utilization increases, the moving average is used only
> > --
> > 2.29.0
> >
>
> Hi,
>
> We do indeed for the diff use the flag for the value updated and no flag for the
> value before the update. However, last_enqueued_diff is only used for the margin
> check which is an heuristic and is not an accurate value (~1%) and as we know
The last_enqueued_diff is compared with the UTIL_EST_MARGIN which is
"1024/100 = 10",
and The LSB may cause ~10% error.
> we already loose the LSB in util_est, I'm not sure this is really necessary.
I'm also not very sure, maybe the calculation will be more rigorous
with the flag?
>
> --
> Vincent
>
On Thu, May 06, 2021 at 08:46:08PM +0800, Xuewen Yan wrote:
> Hi
> On Thu, May 6, 2021 at 8:28 PM Vincent Donnefort
> <[email protected]> wrote:
> >
> > On Thu, May 06, 2021 at 07:09:36PM +0800, Xuewen Yan wrote:
> > > From: Xuewen Yan <[email protected]>
> > >
> > > The UTIL_AVG_UNCHANGED flag had been cleared when the task util changed.
> > > And the enqueued is equal to task_util with the flag, so it is better
> > > to add the UTIL_AVG_UNCHANGED flag for last_enqueued_diff.
Could we change the description here a bit? I don't think this is accurately
explaning the issue. Would probably be interesting to mention that by not
setting the flag, which is the LSB, we add +1 to the diff. This is therefore
reducing slightly UTIL_EST_MARGIN.
> > >
> > > Fixes: b89997aa88f0b sched/pelt: Fix task util_est update filtering
> > >
> > > Signed-off-by: Xuewen Yan <[email protected]>
> > > ---
> > > kernel/sched/fair.c | 2 +-
> > > 1 file changed, 1 insertion(+), 1 deletion(-)
> > >
> > > diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> > > index e5e457fa9dc8..94d77b4fa601 100644
> > > --- a/kernel/sched/fair.c
> > > +++ b/kernel/sched/fair.c
> > > @@ -3996,7 +3996,7 @@ static inline void util_est_update(struct cfs_rq *cfs_rq,
> > > if (ue.enqueued & UTIL_AVG_UNCHANGED)
> > > return;
> > >
> > > - last_enqueued_diff = ue.enqueued;
> > > + last_enqueued_diff = (ue.enqueued | UTIL_AVG_UNCHANGED);
> > >
> > > /*
> > > * Reset EWMA on utilization increases, the moving average is used only
> > > --
> > > 2.29.0
> > >
> >
> > Hi,
> >
> > We do indeed for the diff use the flag for the value updated and no flag for the
> > value before the update. However, last_enqueued_diff is only used for the margin
> > check which is an heuristic and is not an accurate value (~1%) and as we know
> The last_enqueued_diff is compared with the UTIL_EST_MARGIN which is
> "1024/100 = 10",
> and The LSB may cause ~10% error.
I meant ~1% being the original margin. With the bit set, we would use 0.87% instead
of 0.97%.
> > we already loose the LSB in util_est, I'm not sure this is really necessary.
> I'm also not very sure, maybe the calculation will be more rigorous
> with the flag?
> >
> > --
> > Vincent
> >
On Fri, May 7, 2021 at 12:26 AM Vincent Donnefort
<[email protected]> wrote:
>
> On Thu, May 06, 2021 at 08:46:08PM +0800, Xuewen Yan wrote:
> > Hi
> > On Thu, May 6, 2021 at 8:28 PM Vincent Donnefort
> > <[email protected]> wrote:
> > >
> > > On Thu, May 06, 2021 at 07:09:36PM +0800, Xuewen Yan wrote:
> > > > From: Xuewen Yan <[email protected]>
> > > >
> > > > The UTIL_AVG_UNCHANGED flag had been cleared when the task util changed.
> > > > And the enqueued is equal to task_util with the flag, so it is better
> > > > to add the UTIL_AVG_UNCHANGED flag for last_enqueued_diff.
>
> Could we change the description here a bit? I don't think this is accurately
> explaning the issue. Would probably be interesting to mention that by not
> setting the flag, which is the LSB, we add +1 to the diff. This is therefore
> reducing slightly UTIL_EST_MARGIN.
ok, If you agree with this patch, I'll change it in V2.
>
> > > >
> > > > Fixes: b89997aa88f0b sched/pelt: Fix task util_est update filtering
> > > >
> > > > Signed-off-by: Xuewen Yan <[email protected]>
> > > > ---
> > > > kernel/sched/fair.c | 2 +-
> > > > 1 file changed, 1 insertion(+), 1 deletion(-)
> > > >
> > > > diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> > > > index e5e457fa9dc8..94d77b4fa601 100644
> > > > --- a/kernel/sched/fair.c
> > > > +++ b/kernel/sched/fair.c
> > > > @@ -3996,7 +3996,7 @@ static inline void util_est_update(struct cfs_rq *cfs_rq,
> > > > if (ue.enqueued & UTIL_AVG_UNCHANGED)
> > > > return;
> > > >
> > > > - last_enqueued_diff = ue.enqueued;
> > > > + last_enqueued_diff = (ue.enqueued | UTIL_AVG_UNCHANGED);
> > > >
> > > > /*
> > > > * Reset EWMA on utilization increases, the moving average is used only
> > > > --
> > > > 2.29.0
> > > >
> > >
> > > Hi,
> > >
> > > We do indeed for the diff use the flag for the value updated and no flag for the
> > > value before the update. However, last_enqueued_diff is only used for the margin
> > > check which is an heuristic and is not an accurate value (~1%) and as we know
> > The last_enqueued_diff is compared with the UTIL_EST_MARGIN which is
> > "1024/100 = 10",
> > and The LSB may cause ~10% error.
>
> I meant ~1% being the original margin. With the bit set, we would use 0.87% instead
> of 0.97%.
Because the within_margin() does not contain “=”， if the enqueued
without the flag, the margin may be 0.97%(10/1024),
with the flag, be 1.07%(11/1024) instead of 0.87% I think.
>
> > > we already loose the LSB in util_est, I'm not sure this is really necessary.
> > I'm also not very sure, maybe the calculation will be more rigorous
> > with the flag?
> > >
> > > --
> > > Vincent
> > >
On Fri, 7 May 2021 at 03:36, Xuewen Yan <[email protected]> wrote:
>
> On Fri, May 7, 2021 at 12:26 AM Vincent Donnefort
> <[email protected]> wrote:
> >
> > On Thu, May 06, 2021 at 08:46:08PM +0800, Xuewen Yan wrote:
> > > Hi
> > > On Thu, May 6, 2021 at 8:28 PM Vincent Donnefort
> > > <[email protected]> wrote:
> > > >
> > > > On Thu, May 06, 2021 at 07:09:36PM +0800, Xuewen Yan wrote:
> > > > > From: Xuewen Yan <[email protected]>
> > > > >
> > > > > The UTIL_AVG_UNCHANGED flag had been cleared when the task util changed.
> > > > > And the enqueued is equal to task_util with the flag, so it is better
> > > > > to add the UTIL_AVG_UNCHANGED flag for last_enqueued_diff.
> >
> > Could we change the description here a bit? I don't think this is accurately
> > explaning the issue. Would probably be interesting to mention that by not
> > setting the flag, which is the LSB, we add +1 to the diff. This is therefore
> > reducing slightly UTIL_EST_MARGIN.
>
> ok, If you agree with this patch, I'll change it in V2.
Although the impact is not significant , it's worth having an accurate
computation. So the patch makes sense to me. Please submit a v2
> >
> > > > >
> > > > > Fixes: b89997aa88f0b sched/pelt: Fix task util_est update filtering
> > > > >
> > > > > Signed-off-by: Xuewen Yan <[email protected]>
> > > > > ---
> > > > > kernel/sched/fair.c | 2 +-
> > > > > 1 file changed, 1 insertion(+), 1 deletion(-)
> > > > >
> > > > > diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> > > > > index e5e457fa9dc8..94d77b4fa601 100644
> > > > > --- a/kernel/sched/fair.c
> > > > > +++ b/kernel/sched/fair.c
> > > > > @@ -3996,7 +3996,7 @@ static inline void util_est_update(struct cfs_rq *cfs_rq,
> > > > > if (ue.enqueued & UTIL_AVG_UNCHANGED)
> > > > > return;
> > > > >
> > > > > - last_enqueued_diff = ue.enqueued;
> > > > > + last_enqueued_diff = (ue.enqueued | UTIL_AVG_UNCHANGED);
> > > > >
> > > > > /*
> > > > > * Reset EWMA on utilization increases, the moving average is used only
> > > > > --
> > > > > 2.29.0
> > > > >
> > > >
> > > > Hi,
> > > >
> > > > We do indeed for the diff use the flag for the value updated and no flag for the
> > > > value before the update. However, last_enqueued_diff is only used for the margin
> > > > check which is an heuristic and is not an accurate value (~1%) and as we know
> > > The last_enqueued_diff is compared with the UTIL_EST_MARGIN which is
> > > "1024/100 = 10",
> > > and The LSB may cause ~10% error.
> >
> > I meant ~1% being the original margin. With the bit set, we would use 0.87% instead
> > of 0.97%.
>
> Because the within_margin() does not contain “=”， if the enqueued
> without the flag, the margin may be 0.97%(10/1024),
> with the flag, be 1.07%(11/1024) instead of 0.87% I think.
> >
> > > > we already loose the LSB in util_est, I'm not sure this is really necessary.
> > > I'm also not very sure, maybe the calculation will be more rigorous
> > > with the flag?
> > > >
> > > > --
> > > > Vincent
> > > >
On Fri, May 7, 2021 at 2:53 PM Vincent Guittot
<[email protected]> wrote:
>
> On Fri, 7 May 2021 at 03:36, Xuewen Yan <[email protected]> wrote:
> >
> > On Fri, May 7, 2021 at 12:26 AM Vincent Donnefort
> > <[email protected]> wrote:
> > >
> > > On Thu, May 06, 2021 at 08:46:08PM +0800, Xuewen Yan wrote:
> > > > Hi
> > > > On Thu, May 6, 2021 at 8:28 PM Vincent Donnefort
> > > > <[email protected]> wrote:
> > > > >
> > > > > On Thu, May 06, 2021 at 07:09:36PM +0800, Xuewen Yan wrote:
> > > > > > From: Xuewen Yan <[email protected]>
> > > > > >
> > > > > > The UTIL_AVG_UNCHANGED flag had been cleared when the task util changed.
> > > > > > And the enqueued is equal to task_util with the flag, so it is better
> > > > > > to add the UTIL_AVG_UNCHANGED flag for last_enqueued_diff.
> > >
> > > Could we change the description here a bit? I don't think this is accurately
> > > explaning the issue. Would probably be interesting to mention that by not
> > > setting the flag, which is the LSB, we add +1 to the diff. This is therefore
> > > reducing slightly UTIL_EST_MARGIN.
> >
> > ok, If you agree with this patch, I'll change it in V2.
>
> Although the impact is not significant , it's worth having an accurate
> computation. So the patch makes sense to me. Please submit a v2
Okay, I'll submit a v2 later.