lkml.org 
[lkml]   [2019]   [Dec]   [11]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
Date
From
SubjectRe: [PATCH] schied/fair: Skip updating "contrib" without load
On Mon, Dec 09, 2019 at 05:16:27PM +0100, Peter Zijlstra wrote:
> On Sat, Dec 07, 2019 at 12:14:22AM +0800, Peng Wang wrote:
> > We only update load_sum/runnable_load_sum/util_sum with
> > decayed old sum when load is clear.
>
> What you're saying is that because of the:
>
> if (!load)
> runnable = running = 0;
>
> clause in ___update_load_sum(), all the actual users of @contrib in
> accumulate_sum():
>
> if (load)
> sa->load_sum += load * contrib;
> if (runnable)
> sa->runnable_load_sum += runnable * contrib;
> if (running)
> sa->util_sum += contrib << SCHED_CAPACITY_SHIFT;
>
> don't happen, and therefore we don't care what @contrib actually is and
> calculating it is pointless.

Yes.

>
> I suppose that is so. did you happen to have performance numbers? Also,
> I'm thinking this wants a comment.

Actually I don't know how to get the exact performance data.
But I count the times when @load equals zero and not as below:

if (load) {
load_is_not_zero_count++;
contrib = __accumulate_pelt_segments(periods,
1024 - sa->period_contrib, delta);
} else
load_is_zero_count++;

As we can see, load_is_zero_count is much bigger than
load_is_zero_count, and the gap is gradually widening.

load_is_zero_count: 6016044 times
load_is_not_zero_count: 244316 times
19:50:43 up 1 min, 1 user, load average: 0.09, 0.06, 0.02

load_is_zero_count: 7956168 times
load_is_not_zero_count: 261472 times
19:51:42 up 2 min, 1 user, load average: 0.03, 0.05, 0.01

load_is_zero_count: 10199896 times
load_is_not_zero_count: 278364 times
19:52:51 up 3 min, 1 user, load average: 0.06, 0.05, 0.01

load_is_zero_count: 14333700 times
load_is_not_zero_count: 318424 times
19:54:53 up 5 min, 1 user, load average: 0.01, 0.03, 0.00

Perhaps we can gain some performance advantage by saving these unnecessary calculation.

>
> > Signed-off-by: Peng Wang <rocking@linux.alibaba.com>
> > ---
> > kernel/sched/pelt.c | 5 +++--
> > 1 file changed, 3 insertions(+), 2 deletions(-)
> >
> > diff --git a/kernel/sched/pelt.c b/kernel/sched/pelt.c
> > index a96db50..4392953 100644
> > --- a/kernel/sched/pelt.c
> > +++ b/kernel/sched/pelt.c
> > @@ -129,8 +129,9 @@ static u32 __accumulate_pelt_segments(u64 periods, u32 d1, u32 d3)
> > * Step 2
> > */
> > delta %= 1024;
> > - contrib = __accumulate_pelt_segments(periods,
> > - 1024 - sa->period_contrib, delta);
> > + if (load)
> > + contrib = __accumulate_pelt_segments(periods,
> > + 1024 - sa->period_contrib, delta);
> > }
> > sa->period_contrib = delta;
> >
> > --
> > 1.8.3.1
> >

\
 
 \ /
  Last update: 2019-12-11 13:17    [W:0.142 / U:2.040 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site