* [PATCH] sched/fair: fix calc_cfs_shares fixed point arithmetics
@ 2016-12-19 22:40 Samuel Thibault
2016-12-19 22:44 ` Paul Turner
0 siblings, 1 reply; 12+ messages in thread
From: Samuel Thibault @ 2016-12-19 22:40 UTC (permalink / raw)
To: linux-kernel; +Cc: Peter Zijlstra, Thomas Gleixner, Ingo Molnar
2159197d6677 ("sched/core: Enable increased load resolution on 64-bit kernels")
exposed yet another miscalculation in calc_cfs_shares: MIN_SHARES is unscaled,
and must thus be scaled before being manipulated against "shares" amounts.
Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: stable@vger.kernel.org
Fixes: 2159197d6677 ("sched/core: Enable increased load resolution on 64-bit kernels")
---
This should be backported to 4.7 and 4.8 to fix scheduling priorities
miscalculations
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 6559d19..be84f72 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -2657,8 +2657,8 @@ static long calc_cfs_shares(struct cfs_rq *cfs_rq, struct task_group *tg)
if (tg_weight)
shares /= tg_weight;
- if (shares < MIN_SHARES)
- shares = MIN_SHARES;
+ if (shares < scale_load(MIN_SHARES))
+ shares = scale_load(MIN_SHARES);
if (shares > tg->shares)
shares = tg->shares;
^ permalink raw reply related [flat|nested] 12+ messages in thread
* Re: [PATCH] sched/fair: fix calc_cfs_shares fixed point arithmetics
2016-12-19 22:40 [PATCH] sched/fair: fix calc_cfs_shares fixed point arithmetics Samuel Thibault
@ 2016-12-19 22:44 ` Paul Turner
2016-12-19 23:07 ` Samuel Thibault
0 siblings, 1 reply; 12+ messages in thread
From: Paul Turner @ 2016-12-19 22:44 UTC (permalink / raw)
To: Samuel Thibault, LKML, Peter Zijlstra, Thomas Gleixner, Ingo Molnar
On Mon, Dec 19, 2016 at 2:40 PM, Samuel Thibault
<samuel.thibault@ens-lyon.org> wrote:
> 2159197d6677 ("sched/core: Enable increased load resolution on 64-bit kernels")
>
> exposed yet another miscalculation in calc_cfs_shares: MIN_SHARES is unscaled,
> and must thus be scaled before being manipulated against "shares" amounts.
It's actually intentional that MIN_SHARES is un-scaled here, this is
necessary to support the goal of sub-partitioning groups with small
shares.
E.g. A group with shares=2 and 5 threads will internally provide 2048
units of weight for the load-balancer to account for their
distribution.
>
> Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
> Cc: Peter Zijlstra <peterz@infradead.org>
> Cc: Thomas Gleixner <tglx@linutronix.de>
> Cc: Ingo Molnar <mingo@kernel.org>
> Cc: stable@vger.kernel.org
> Fixes: 2159197d6677 ("sched/core: Enable increased load resolution on 64-bit kernels")
>
> ---
> This should be backported to 4.7 and 4.8 to fix scheduling priorities
> miscalculations
>
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index 6559d19..be84f72 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -2657,8 +2657,8 @@ static long calc_cfs_shares(struct cfs_rq *cfs_rq, struct task_group *tg)
> if (tg_weight)
> shares /= tg_weight;
>
> - if (shares < MIN_SHARES)
> - shares = MIN_SHARES;
> + if (shares < scale_load(MIN_SHARES))
> + shares = scale_load(MIN_SHARES);
> if (shares > tg->shares)
> shares = tg->shares;
>
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [PATCH] sched/fair: fix calc_cfs_shares fixed point arithmetics
2016-12-19 22:44 ` Paul Turner
@ 2016-12-19 23:07 ` Samuel Thibault
2016-12-19 23:26 ` Paul Turner
0 siblings, 1 reply; 12+ messages in thread
From: Samuel Thibault @ 2016-12-19 23:07 UTC (permalink / raw)
To: Paul Turner; +Cc: LKML, Peter Zijlstra, Thomas Gleixner, Ingo Molnar
Paul Turner, on Mon 19 Dec 2016 14:44:38 -0800, wrote:
> On Mon, Dec 19, 2016 at 2:40 PM, Samuel Thibault
> <samuel.thibault@ens-lyon.org> wrote:
> > 2159197d6677 ("sched/core: Enable increased load resolution on 64-bit kernels")
> >
> > exposed yet another miscalculation in calc_cfs_shares: MIN_SHARES is unscaled,
> > and must thus be scaled before being manipulated against "shares" amounts.
>
> It's actually intentional that MIN_SHARES is un-scaled here, this is
> necessary to support the goal of sub-partitioning groups with small
> shares.
Uh? you mean it's normal that MIN_SHARES is here compared as such
against "shares" while e.g. in sched_group_set_shares or effective_load
it is scaled before comparing with "shares"?
> E.g. A group with shares=2 and 5 threads will internally provide 2048
> units of weight for the load-balancer to account for their
> distribution.
But here "shares" is already scaled, so
> > - if (shares < MIN_SHARES)
> > - shares = MIN_SHARES;
...
> > return shares;
This will only make sure that the returned shares is 2, not 2048.
Samuel
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [PATCH] sched/fair: fix calc_cfs_shares fixed point arithmetics
2016-12-19 23:07 ` Samuel Thibault
@ 2016-12-19 23:26 ` Paul Turner
2016-12-19 23:29 ` Samuel Thibault
0 siblings, 1 reply; 12+ messages in thread
From: Paul Turner @ 2016-12-19 23:26 UTC (permalink / raw)
To: Samuel Thibault, Paul Turner, LKML, Peter Zijlstra,
Thomas Gleixner, Ingo Molnar
On Mon, Dec 19, 2016 at 3:07 PM, Samuel Thibault
<samuel.thibault@ens-lyon.org> wrote:
> Paul Turner, on Mon 19 Dec 2016 14:44:38 -0800, wrote:
>> On Mon, Dec 19, 2016 at 2:40 PM, Samuel Thibault
>> <samuel.thibault@ens-lyon.org> wrote:
>> > 2159197d6677 ("sched/core: Enable increased load resolution on 64-bit kernels")
>> >
>> > exposed yet another miscalculation in calc_cfs_shares: MIN_SHARES is unscaled,
>> > and must thus be scaled before being manipulated against "shares" amounts.
>>
>> It's actually intentional that MIN_SHARES is un-scaled here, this is
>> necessary to support the goal of sub-partitioning groups with small
>> shares.
>
> Uh? you mean it's normal that MIN_SHARES is here compared as such
> against "shares" while e.g. in sched_group_set_shares or effective_load
> it is scaled before comparing with "shares"?
Yes.
sched_group_set_shares() controls the amount allocated to the group.
Both calc_cfs_shares() and effective_load() are subdividing this
total. Which is why it is scaled up from the external value of 2.
>
>> E.g. A group with shares=2 and 5 threads will internally provide 2048
>> units of weight for the load-balancer to account for their
>> distribution.
>
> But here "shares" is already scaled, so
>
>> > - if (shares < MIN_SHARES)
>> > - shares = MIN_SHARES;
> ...
>> > return shares;
>
> This will only make sure that the returned shares is 2, not 2048.
This is intentional. The MIN_SHARES you are seeing here is overloaded.
Every "1" unit of share is "SCHED_LOAD_RESOLUTION" bits internally.
We express a minimum of "2" in terms of the unit weight due to larger
numerical errors in the "1" case.
In the unscaled case this needs to be MIN_SHARES, and in the scaled
case, the subdivision of the scaled values must still be >=2.
To make this concrete:
In this case we can then internally say that there are (internally)
~410 units of weight for each of these 5 threads.
Thus, if one cpu has 4 threads and another 1, we see that as a
1640/410 imbalance, not a 2048/2048 balance.
>
> Samuel
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [PATCH] sched/fair: fix calc_cfs_shares fixed point arithmetics
2016-12-19 23:26 ` Paul Turner
@ 2016-12-19 23:29 ` Samuel Thibault
2016-12-19 23:32 ` Paul Turner
0 siblings, 1 reply; 12+ messages in thread
From: Samuel Thibault @ 2016-12-19 23:29 UTC (permalink / raw)
To: Paul Turner; +Cc: LKML, Peter Zijlstra, Thomas Gleixner, Ingo Molnar
Paul Turner, on Mon 19 Dec 2016 15:26:19 -0800, wrote:
> >> > - if (shares < MIN_SHARES)
> >> > - shares = MIN_SHARES;
> > ...
> >> > return shares;
> >
> > This will only make sure that the returned shares is 2, not 2048.
>
> This is intentional. The MIN_SHARES you are seeing here is overloaded.
> Every "1" unit of share is "SCHED_LOAD_RESOLUTION" bits internally.
I'm not talking about the SCHED_LOAD_RESOLUTION scaling, but about the
SCHED_FIXEDPOINT_SHIFT scaling, which is what
2159197d6677 ("sched/core: Enable increased load resolution on 64-bit kernels")
modified on 64bit platforms.
Samuel
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [PATCH] sched/fair: fix calc_cfs_shares fixed point arithmetics
2016-12-19 23:29 ` Samuel Thibault
@ 2016-12-19 23:32 ` Paul Turner
2016-12-19 23:45 ` Samuel Thibault
0 siblings, 1 reply; 12+ messages in thread
From: Paul Turner @ 2016-12-19 23:32 UTC (permalink / raw)
To: Samuel Thibault, Paul Turner, LKML, Peter Zijlstra,
Thomas Gleixner, Ingo Molnar
On Mon, Dec 19, 2016 at 3:29 PM, Samuel Thibault
<samuel.thibault@ens-lyon.org> wrote:
> Paul Turner, on Mon 19 Dec 2016 15:26:19 -0800, wrote:
>> >> > - if (shares < MIN_SHARES)
>> >> > - shares = MIN_SHARES;
>> > ...
>> >> > return shares;
>> >
>> > This will only make sure that the returned shares is 2, not 2048.
>>
>> This is intentional. The MIN_SHARES you are seeing here is overloaded.
>> Every "1" unit of share is "SCHED_LOAD_RESOLUTION" bits internally.
>
> I'm not talking about the SCHED_LOAD_RESOLUTION scaling, but about the
> SCHED_FIXEDPOINT_SHIFT scaling, which is what
> 2159197d6677 ("sched/core: Enable increased load resolution on 64-bit kernels")
> modified on 64bit platforms.
.... From that commit:
"""
-#if 0 /* BITS_PER_LONG > 32 -- currently broken: it increases power
usage under light load */
+#ifdef CONFIG_64BIT
# define SCHED_LOAD_RESOLUTION 10
# define scale_load(w) ((w) << SCHED_LOAD_RESOLUTION)
# define scale_load_down(w) ((w) >> SCHED_LOAD_RESOLUTION)
"""
Please take a deeper look at the scale_load() interactions.
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [PATCH] sched/fair: fix calc_cfs_shares fixed point arithmetics
2016-12-19 23:32 ` Paul Turner
@ 2016-12-19 23:45 ` Samuel Thibault
2016-12-20 13:04 ` Dietmar Eggemann
0 siblings, 1 reply; 12+ messages in thread
From: Samuel Thibault @ 2016-12-19 23:45 UTC (permalink / raw)
To: Paul Turner; +Cc: LKML, Peter Zijlstra, Thomas Gleixner, Ingo Molnar
Paul Turner, on Mon 19 Dec 2016 15:32:15 -0800, wrote:
> On Mon, Dec 19, 2016 at 3:29 PM, Samuel Thibault
> <samuel.thibault@ens-lyon.org> wrote:
> > Paul Turner, on Mon 19 Dec 2016 15:26:19 -0800, wrote:
> >> >> > - if (shares < MIN_SHARES)
> >> >> > - shares = MIN_SHARES;
> >> > ...
> >> >> > return shares;
> >> >
> >> > This will only make sure that the returned shares is 2, not 2048.
> >>
> >> This is intentional. The MIN_SHARES you are seeing here is overloaded.
> >> Every "1" unit of share is "SCHED_LOAD_RESOLUTION" bits internally.
> >
> > I'm not talking about the SCHED_LOAD_RESOLUTION scaling, but about the
> > SCHED_FIXEDPOINT_SHIFT scaling, which is what
> > 2159197d6677 ("sched/core: Enable increased load resolution on 64-bit kernels")
> > modified on 64bit platforms.
>
> .... From that commit:
>
> """
> -#if 0 /* BITS_PER_LONG > 32 -- currently broken: it increases power
> usage under light load */
> +#ifdef CONFIG_64BIT
> # define SCHED_LOAD_RESOLUTION 10
> # define scale_load(w) ((w) << SCHED_LOAD_RESOLUTION)
> # define scale_load_down(w) ((w) >> SCHED_LOAD_RESOLUTION)
Errgl, sorry, I was referring to the old naming. This stuff has seen
so much patching over and over in the past revisions... It though you
were referring to SCHED_CAPACITY_SCALE. The code I was reading now uses
SCHED_LOAD_RESOLUTION, so that's why I read your "SCHED_LOAD_RESOLUTION"
as "the other scaling".
> The MIN_SHARES you are seeing here is overloaded.
> In the unscaled case this needs to be MIN_SHARES, and in the scaled
> case, the subdivision of the scaled values must still be >=2.
Ok, now I understand. I have to say this overloading is confusing.
Samuel
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [PATCH] sched/fair: fix calc_cfs_shares fixed point arithmetics
2016-12-19 23:45 ` Samuel Thibault
@ 2016-12-20 13:04 ` Dietmar Eggemann
2016-12-20 13:15 ` Peter Zijlstra
2016-12-20 13:22 ` [PATCH] sched/fair: fix calc_cfs_shares fixed point arithmetics Samuel Thibault
0 siblings, 2 replies; 12+ messages in thread
From: Dietmar Eggemann @ 2016-12-20 13:04 UTC (permalink / raw)
To: Samuel Thibault, Paul Turner, LKML, Peter Zijlstra,
Thomas Gleixner, Ingo Molnar
Hi Samuel,
On 12/20/2016 12:45 AM, Samuel Thibault wrote:
> Paul Turner, on Mon 19 Dec 2016 15:32:15 -0800, wrote:
>> On Mon, Dec 19, 2016 at 3:29 PM, Samuel Thibault
>> <samuel.thibault@ens-lyon.org> wrote:
>>> Paul Turner, on Mon 19 Dec 2016 15:26:19 -0800, wrote:
[...]
>> The MIN_SHARES you are seeing here is overloaded.
>> In the unscaled case this needs to be MIN_SHARES, and in the scaled
>> case, the subdivision of the scaled values must still be >=2.
>
> Ok, now I understand. I have to say this overloading is confusing.
>
> Samuel
this had been already discussed back in August when I posted the
original patch.
https://lkml.org/lkml/2016/8/22/351
https://lkml.org/lkml/2016/8/23/641
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [PATCH] sched/fair: fix calc_cfs_shares fixed point arithmetics
2016-12-20 13:04 ` Dietmar Eggemann
@ 2016-12-20 13:15 ` Peter Zijlstra
2017-01-11 11:29 ` Dietmar Eggemann
2016-12-20 13:22 ` [PATCH] sched/fair: fix calc_cfs_shares fixed point arithmetics Samuel Thibault
1 sibling, 1 reply; 12+ messages in thread
From: Peter Zijlstra @ 2016-12-20 13:15 UTC (permalink / raw)
To: Dietmar Eggemann
Cc: Samuel Thibault, Paul Turner, LKML, Thomas Gleixner, Ingo Molnar
On Tue, Dec 20, 2016 at 02:04:34PM +0100, Dietmar Eggemann wrote:
> Hi Samuel,
>
> On 12/20/2016 12:45 AM, Samuel Thibault wrote:
> >Paul Turner, on Mon 19 Dec 2016 15:32:15 -0800, wrote:
> >>On Mon, Dec 19, 2016 at 3:29 PM, Samuel Thibault
> >><samuel.thibault@ens-lyon.org> wrote:
> >>>Paul Turner, on Mon 19 Dec 2016 15:26:19 -0800, wrote:
>
> [...]
>
> >>The MIN_SHARES you are seeing here is overloaded.
> >>In the unscaled case this needs to be MIN_SHARES, and in the scaled
> >>case, the subdivision of the scaled values must still be >=2.
> >
> >Ok, now I understand. I have to say this overloading is confusing.
> >
> >Samuel
>
> this had been already discussed back in August when I posted the original
> patch.
Maybe we should put a comment in to avoid getting more of these ;-)
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [PATCH] sched/fair: fix calc_cfs_shares fixed point arithmetics
2016-12-20 13:04 ` Dietmar Eggemann
2016-12-20 13:15 ` Peter Zijlstra
@ 2016-12-20 13:22 ` Samuel Thibault
1 sibling, 0 replies; 12+ messages in thread
From: Samuel Thibault @ 2016-12-20 13:22 UTC (permalink / raw)
To: Dietmar Eggemann
Cc: Paul Turner, LKML, Peter Zijlstra, Thomas Gleixner, Ingo Molnar
Dietmar Eggemann, on Tue 20 Dec 2016 14:04:34 +0100, wrote:
> On 12/20/2016 12:45 AM, Samuel Thibault wrote:
> >Paul Turner, on Mon 19 Dec 2016 15:32:15 -0800, wrote:
> >>On Mon, Dec 19, 2016 at 3:29 PM, Samuel Thibault
> >><samuel.thibault@ens-lyon.org> wrote:
> >>>Paul Turner, on Mon 19 Dec 2016 15:26:19 -0800, wrote:
>
> [...]
>
> >>The MIN_SHARES you are seeing here is overloaded.
> >>In the unscaled case this needs to be MIN_SHARES, and in the scaled
> >>case, the subdivision of the scaled values must still be >=2.
> >
> >Ok, now I understand. I have to say this overloading is confusing.
> >
> >Samuel
>
> this had been already discussed back in August when I posted the original
> patch.
But that doesn't show up in the source code or git history. One
shouldn't have to dig mailing lists to get code comments :)
Samuel
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [PATCH] sched/fair: fix calc_cfs_shares fixed point arithmetics
2016-12-20 13:15 ` Peter Zijlstra
@ 2017-01-11 11:29 ` Dietmar Eggemann
2017-01-14 12:48 ` [tip:sched/core] sched/fair: Explain why MIN_SHARES isn't scaled in calc_cfs_shares() tip-bot for Dietmar Eggemann
0 siblings, 1 reply; 12+ messages in thread
From: Dietmar Eggemann @ 2017-01-11 11:29 UTC (permalink / raw)
To: Peter Zijlstra
Cc: Samuel Thibault, Paul Turner, LKML, Thomas Gleixner, Ingo Molnar
On 20/12/16 13:15, Peter Zijlstra wrote:
> On Tue, Dec 20, 2016 at 02:04:34PM +0100, Dietmar Eggemann wrote:
>> Hi Samuel,
>>
>> On 12/20/2016 12:45 AM, Samuel Thibault wrote:
>>> Paul Turner, on Mon 19 Dec 2016 15:32:15 -0800, wrote:
>>>> On Mon, Dec 19, 2016 at 3:29 PM, Samuel Thibault
>>>> <samuel.thibault@ens-lyon.org> wrote:
>>>>> Paul Turner, on Mon 19 Dec 2016 15:26:19 -0800, wrote:
>>
>> [...]
>>
>>>> The MIN_SHARES you are seeing here is overloaded.
>>>> In the unscaled case this needs to be MIN_SHARES, and in the scaled
>>>> case, the subdivision of the scaled values must still be >=2.
>>>
>>> Ok, now I understand. I have to say this overloading is confusing.
>>>
>>> Samuel
>>
>> this had been already discussed back in August when I posted the original
>> patch.
>
> Maybe we should put a comment in to avoid getting more of these ;-)
>
Maybe something like this? Mainly what Paul taught us plus an example from
a discussion I had with Vincent.
-- >8 --
Subject: [PATCH] sched/fair: Explain why MIN_SHARES isn't scaled in
calc_cfs_shares()
Signed-off-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
---
kernel/sched/fair.c | 12 ++++++++++++
1 file changed, 12 insertions(+)
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 6559d197e08a..a0ca9b11b1b3 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -2657,6 +2657,18 @@ static long calc_cfs_shares(struct cfs_rq *cfs_rq, struct task_group *tg)
if (tg_weight)
shares /= tg_weight;
+ /*
+ * MIN_SHARES has to be unscaled here to support per-cpu partitioning
+ * of a group with small tg->shares value. It is a floor value which is
+ * assigned as a minimum load.weight to the sched_entity representing
+ * the group on a cpu.
+ *
+ * E.g. on 64-bit for a group with tg->shares of scale_load(15)=15*1024
+ * on an 8 core system with 8 tasks each runnable on one cpu shares has
+ * to be 15*1024*1/8=1920 instead of scale_load(MIN_SHARES)=2*1024. In
+ * case no task is runnable on a cpu MIN_SHARES=2 should be returned
+ * instead of 0.
+ */
if (shares < MIN_SHARES)
shares = MIN_SHARES;
if (shares > tg->shares)
--
2.11.0
^ permalink raw reply related [flat|nested] 12+ messages in thread
* [tip:sched/core] sched/fair: Explain why MIN_SHARES isn't scaled in calc_cfs_shares()
2017-01-11 11:29 ` Dietmar Eggemann
@ 2017-01-14 12:48 ` tip-bot for Dietmar Eggemann
0 siblings, 0 replies; 12+ messages in thread
From: tip-bot for Dietmar Eggemann @ 2017-01-14 12:48 UTC (permalink / raw)
To: linux-tip-commits
Cc: tglx, dietmar.eggemann, pjt, hpa, efault, samuel.thibault, mingo,
torvalds, linux-kernel, peterz
Commit-ID: b8fd8423697b9ec729c5bb91737faad84ae19985
Gitweb: http://git.kernel.org/tip/b8fd8423697b9ec729c5bb91737faad84ae19985
Author: Dietmar Eggemann <dietmar.eggemann@arm.com>
AuthorDate: Wed, 11 Jan 2017 11:29:47 +0000
Committer: Ingo Molnar <mingo@kernel.org>
CommitDate: Sat, 14 Jan 2017 11:30:02 +0100
sched/fair: Explain why MIN_SHARES isn't scaled in calc_cfs_shares()
Signed-off-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mike Galbraith <efault@gmx.de>
Cc: Paul Turner <pjt@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Samuel Thibault <samuel.thibault@ens-lyon.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/e9a4d858-bcf3-36b9-e3a9-449953e34569@arm.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
---
kernel/sched/fair.c | 12 ++++++++++++
1 file changed, 12 insertions(+)
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 2b866a2..274c747 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -2657,6 +2657,18 @@ static long calc_cfs_shares(struct cfs_rq *cfs_rq, struct task_group *tg)
if (tg_weight)
shares /= tg_weight;
+ /*
+ * MIN_SHARES has to be unscaled here to support per-CPU partitioning
+ * of a group with small tg->shares value. It is a floor value which is
+ * assigned as a minimum load.weight to the sched_entity representing
+ * the group on a CPU.
+ *
+ * E.g. on 64-bit for a group with tg->shares of scale_load(15)=15*1024
+ * on an 8-core system with 8 tasks each runnable on one CPU shares has
+ * to be 15*1024*1/8=1920 instead of scale_load(MIN_SHARES)=2*1024. In
+ * case no task is runnable on a CPU MIN_SHARES=2 should be returned
+ * instead of 0.
+ */
if (shares < MIN_SHARES)
shares = MIN_SHARES;
if (shares > tg->shares)
^ permalink raw reply related [flat|nested] 12+ messages in thread
end of thread, other threads:[~2017-01-14 12:49 UTC | newest]
Thread overview: 12+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2016-12-19 22:40 [PATCH] sched/fair: fix calc_cfs_shares fixed point arithmetics Samuel Thibault
2016-12-19 22:44 ` Paul Turner
2016-12-19 23:07 ` Samuel Thibault
2016-12-19 23:26 ` Paul Turner
2016-12-19 23:29 ` Samuel Thibault
2016-12-19 23:32 ` Paul Turner
2016-12-19 23:45 ` Samuel Thibault
2016-12-20 13:04 ` Dietmar Eggemann
2016-12-20 13:15 ` Peter Zijlstra
2017-01-11 11:29 ` Dietmar Eggemann
2017-01-14 12:48 ` [tip:sched/core] sched/fair: Explain why MIN_SHARES isn't scaled in calc_cfs_shares() tip-bot for Dietmar Eggemann
2016-12-20 13:22 ` [PATCH] sched/fair: fix calc_cfs_shares fixed point arithmetics Samuel Thibault
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).