* [PATCH v3] sched/fair: Use all little CPUs for CPU-bound workload
@ 2023-12-06 9:00 Pierre Gondois
2023-12-07 17:22 ` Qais Yousef
2023-12-23 16:09 ` [tip: sched/core] sched/fair: Use all little CPUs for CPU-bound workloads tip-bot2 for Pierre Gondois
0 siblings, 2 replies; 3+ messages in thread
From: Pierre Gondois @ 2023-12-06 9:00 UTC (permalink / raw)
To: linux-kernel
Cc: Qais Yousef, Pierre Gondois, Vincent Guittot, Dietmar Eggemann,
Ingo Molnar, Peter Zijlstra, Juri Lelli, Steven Rostedt,
Ben Segall, Mel Gorman, Daniel Bristot de Oliveira,
Valentin Schneider
Running n CPU-bound tasks on an n CPUs platform:
- with asymmetric CPU capacity
- not being a DynamIq system (i.e. having a PKG level sched domain
without the SD_SHARE_PKG_RESOURCES flag set)
might result in a task placement where two tasks run on a big CPU
and none on a little CPU. This placement could be more optimal by
using all CPUs.
Testing platform:
Juno-r2:
- 2 big CPUs (1-2), maximum capacity of 1024
- 4 little CPUs (0,3-5), maximum capacity of 383
Testing workload ([1]):
Spawn 6 CPU-bound tasks. During the first 100ms (step 1), each tasks
is affine to a CPU, except for:
- one little CPU which is left idle.
- one big CPU which has 2 tasks affine.
After the 100ms (step 2), remove the cpumask affinity.
Before patch:
During step 2, the load balancer running from the idle CPU tags sched
domains as:
- little CPUs: 'group_has_spare'. Cf. group_has_capacity() and
group_is_overloaded(), 3 CPU-bound tasks run on a 4 CPUs
sched-domain, and the idle CPU provides enough spare capacity
regarding the imbalance_pct
- big CPUs: 'group_overloaded'. Indeed, 3 tasks run on a 2 CPUs
sched-domain, so the following path is used:
group_is_overloaded()
\-if (sgs->sum_nr_running <= sgs->group_weight) return true;
The following path which would change the migration type to
'migrate_task' is not taken:
calculate_imbalance()
\-if (env->idle != CPU_NOT_IDLE && env->imbalance == 0)
as the local group has some spare capacity, so the imbalance
is not 0.
The migration type requested is 'migrate_util' and the busiest
runqueue is the big CPU's runqueue having 2 tasks (each having a
utilization of 512). The idle little CPU cannot pull one of these
task as its capacity is too small for the task. The following path
is used:
detach_tasks()
\-case migrate_util:
\-if (util > env->imbalance) goto next;
After patch:
As the number of failed balancing attempts grows (with
'nr_balance_failed'), progressively make it easier to migrate
a big task to the idling little CPU. A similar mechanism is
used for the 'migrate_load' migration type.
Improvement:
Running the testing workload [1] with the step 2 representing
a ~10s load for a big CPU:
Before patch: ~19.3s
After patch: ~18s (-6.7%)
Similar issue reported at:
https://lore.kernel.org/lkml/20230716014125.139577-1-qyousef@layalina.io/
v1:
https://lore.kernel.org/all/20231110125902.2152380-1-pierre.gondois@arm.com/
v2:
https://lore.kernel.org/all/20231124153323.3202444-1-pierre.gondois@arm.com/
Suggested-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Pierre Gondois <pierre.gondois@arm.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
---
Notes:
v2:
- Used Vincent's approach.
v3:
- Updated commit message.
- Added Reviewed-by tags
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 d7a3c63a2171..9481b8cff31b 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -9060,7 +9060,7 @@ static int detach_tasks(struct lb_env *env)
case migrate_util:
util = task_util_est(p);
- if (util > env->imbalance)
+ if (shr_bound(util, env->sd->nr_balance_failed) > env->imbalance)
goto next;
env->imbalance -= util;
--
2.25.1
^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [PATCH v3] sched/fair: Use all little CPUs for CPU-bound workload
2023-12-06 9:00 [PATCH v3] sched/fair: Use all little CPUs for CPU-bound workload Pierre Gondois
@ 2023-12-07 17:22 ` Qais Yousef
2023-12-23 16:09 ` [tip: sched/core] sched/fair: Use all little CPUs for CPU-bound workloads tip-bot2 for Pierre Gondois
1 sibling, 0 replies; 3+ messages in thread
From: Qais Yousef @ 2023-12-07 17:22 UTC (permalink / raw)
To: Pierre Gondois
Cc: linux-kernel, Vincent Guittot, Dietmar Eggemann, Ingo Molnar,
Peter Zijlstra, Juri Lelli, Steven Rostedt, Ben Segall,
Mel Gorman, Daniel Bristot de Oliveira, Valentin Schneider
On 12/06/23 10:00, Pierre Gondois wrote:
> Running n CPU-bound tasks on an n CPUs platform:
> - with asymmetric CPU capacity
> - not being a DynamIq system (i.e. having a PKG level sched domain
> without the SD_SHARE_PKG_RESOURCES flag set)
> might result in a task placement where two tasks run on a big CPU
> and none on a little CPU. This placement could be more optimal by
> using all CPUs.
>
> Testing platform:
> Juno-r2:
> - 2 big CPUs (1-2), maximum capacity of 1024
> - 4 little CPUs (0,3-5), maximum capacity of 383
>
> Testing workload ([1]):
> Spawn 6 CPU-bound tasks. During the first 100ms (step 1), each tasks
> is affine to a CPU, except for:
> - one little CPU which is left idle.
> - one big CPU which has 2 tasks affine.
> After the 100ms (step 2), remove the cpumask affinity.
>
> Before patch:
> During step 2, the load balancer running from the idle CPU tags sched
> domains as:
> - little CPUs: 'group_has_spare'. Cf. group_has_capacity() and
> group_is_overloaded(), 3 CPU-bound tasks run on a 4 CPUs
> sched-domain, and the idle CPU provides enough spare capacity
> regarding the imbalance_pct
> - big CPUs: 'group_overloaded'. Indeed, 3 tasks run on a 2 CPUs
> sched-domain, so the following path is used:
> group_is_overloaded()
> \-if (sgs->sum_nr_running <= sgs->group_weight) return true;
>
> The following path which would change the migration type to
> 'migrate_task' is not taken:
> calculate_imbalance()
> \-if (env->idle != CPU_NOT_IDLE && env->imbalance == 0)
> as the local group has some spare capacity, so the imbalance
> is not 0.
>
> The migration type requested is 'migrate_util' and the busiest
> runqueue is the big CPU's runqueue having 2 tasks (each having a
> utilization of 512). The idle little CPU cannot pull one of these
> task as its capacity is too small for the task. The following path
> is used:
> detach_tasks()
> \-case migrate_util:
> \-if (util > env->imbalance) goto next;
>
> After patch:
> As the number of failed balancing attempts grows (with
> 'nr_balance_failed'), progressively make it easier to migrate
> a big task to the idling little CPU. A similar mechanism is
> used for the 'migrate_load' migration type.
>
> Improvement:
> Running the testing workload [1] with the step 2 representing
> a ~10s load for a big CPU:
> Before patch: ~19.3s
> After patch: ~18s (-6.7%)
>
> Similar issue reported at:
> https://lore.kernel.org/lkml/20230716014125.139577-1-qyousef@layalina.io/
>
> v1:
> https://lore.kernel.org/all/20231110125902.2152380-1-pierre.gondois@arm.com/
> v2:
> https://lore.kernel.org/all/20231124153323.3202444-1-pierre.gondois@arm.com/
>
> Suggested-by: Vincent Guittot <vincent.guittot@linaro.org>
> Signed-off-by: Pierre Gondois <pierre.gondois@arm.com>
> Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
> Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
> ---
Thanks Pierre! I think this is a good candidate to stable. It is likely to help
some folks shipping with phantom domains on impacted stable kernels.
Cheers
--
Qais Yousef
^ permalink raw reply [flat|nested] 3+ messages in thread
* [tip: sched/core] sched/fair: Use all little CPUs for CPU-bound workloads
2023-12-06 9:00 [PATCH v3] sched/fair: Use all little CPUs for CPU-bound workload Pierre Gondois
2023-12-07 17:22 ` Qais Yousef
@ 2023-12-23 16:09 ` tip-bot2 for Pierre Gondois
1 sibling, 0 replies; 3+ messages in thread
From: tip-bot2 for Pierre Gondois @ 2023-12-23 16:09 UTC (permalink / raw)
To: linux-tip-commits
Cc: Vincent Guittot, Pierre Gondois, Ingo Molnar, Dietmar Eggemann,
Qais Yousef, x86, linux-kernel
The following commit has been merged into the sched/core branch of tip:
Commit-ID: 3af7524b14198f5159a86692d57a9f28ec9375ce
Gitweb: https://git.kernel.org/tip/3af7524b14198f5159a86692d57a9f28ec9375ce
Author: Pierre Gondois <pierre.gondois@arm.com>
AuthorDate: Wed, 06 Dec 2023 10:00:43 +01:00
Committer: Ingo Molnar <mingo@kernel.org>
CommitterDate: Sat, 23 Dec 2023 16:06:36 +01:00
sched/fair: Use all little CPUs for CPU-bound workloads
Running N CPU-bound tasks on an N CPUs platform:
- with asymmetric CPU capacity
- not being a DynamIq system (i.e. having a PKG level sched domain
without the SD_SHARE_PKG_RESOURCES flag set)
.. might result in a task placement where two tasks run on a big CPU
and none on a little CPU. This placement could be more optimal by
using all CPUs.
Testing platform:
Juno-r2:
- 2 big CPUs (1-2), maximum capacity of 1024
- 4 little CPUs (0,3-5), maximum capacity of 383
Testing workload ([1]):
Spawn 6 CPU-bound tasks. During the first 100ms (step 1), each tasks
is affine to a CPU, except for:
- one little CPU which is left idle.
- one big CPU which has 2 tasks affine.
After the 100ms (step 2), remove the cpumask affinity.
Behavior before the patch:
During step 2, the load balancer running from the idle CPU tags sched
domains as:
- little CPUs: 'group_has_spare'. Cf. group_has_capacity() and
group_is_overloaded(), 3 CPU-bound tasks run on a 4 CPUs
sched-domain, and the idle CPU provides enough spare capacity
regarding the imbalance_pct
- big CPUs: 'group_overloaded'. Indeed, 3 tasks run on a 2 CPUs
sched-domain, so the following path is used:
group_is_overloaded()
\-if (sgs->sum_nr_running <= sgs->group_weight) return true;
The following path which would change the migration type to
'migrate_task' is not taken:
calculate_imbalance()
\-if (env->idle != CPU_NOT_IDLE && env->imbalance == 0)
as the local group has some spare capacity, so the imbalance
is not 0.
The migration type requested is 'migrate_util' and the busiest
runqueue is the big CPU's runqueue having 2 tasks (each having a
utilization of 512). The idle little CPU cannot pull one of these
task as its capacity is too small for the task. The following path
is used:
detach_tasks()
\-case migrate_util:
\-if (util > env->imbalance) goto next;
After the patch:
As the number of failed balancing attempts grows (with
'nr_balance_failed'), progressively make it easier to migrate
a big task to the idling little CPU. A similar mechanism is
used for the 'migrate_load' migration type.
Improvement:
Running the testing workload [1] with the step 2 representing
a ~10s load for a big CPU:
Before patch: ~19.3s
After patch: ~18s (-6.7%)
Similar issue reported at:
https://lore.kernel.org/lkml/20230716014125.139577-1-qyousef@layalina.io/
Suggested-by: Vincent Guittot <vincent.guittot@linaro.org>
Signed-off-by: Pierre Gondois <pierre.gondois@arm.com>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Dietmar Eggemann <dietmar.eggemann@arm.com>
Acked-by: Qais Yousef <qyousef@layalina.io>
Link: https://lore.kernel.org/r/20231206090043.634697-1-pierre.gondois@arm.com
---
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 823dd76..1d561b5 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -9071,7 +9071,7 @@ static int detach_tasks(struct lb_env *env)
case migrate_util:
util = task_util_est(p);
- if (util > env->imbalance)
+ if (shr_bound(util, env->sd->nr_balance_failed) > env->imbalance)
goto next;
env->imbalance -= util;
^ permalink raw reply related [flat|nested] 3+ messages in thread
end of thread, other threads:[~2023-12-23 16:09 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-12-06 9:00 [PATCH v3] sched/fair: Use all little CPUs for CPU-bound workload Pierre Gondois
2023-12-07 17:22 ` Qais Yousef
2023-12-23 16:09 ` [tip: sched/core] sched/fair: Use all little CPUs for CPU-bound workloads tip-bot2 for Pierre Gondois
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).