[PATCH v2] sched/fair: Optimize select_idle_cpu

[PATCH v2] sched/fair: Optimize select_idle_cpu

[Cheng Jian]

select_idle_cpu() will scan the LLC domain for idle CPUs,
it's always expensive. so the next commit :

	1ad3aaf3fcd2 ("sched/core: Implement new approach to scale select_idle_cpu()")

introduces a way to limit how many CPUs we scan.

But it consume some CPUs out of 'nr' that are not allowed
for the task and thus waste our attempts. The function
always return nr_cpumask_bits, and we can't find a CPU
which our task is allowed to run.

Cpumask may be too big, similar to select_idle_core(), use
per_cpu_ptr 'select_idle_mask' to prevent stack overflow.

Fixes: 1ad3aaf3fcd2 ("sched/core: Implement new approach to scale select_idle_cpu()")
Signed-off-by: Cheng Jian <cj.chengjian@huawei.com>
---
 kernel/sched/fair.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 08a233e97a01..d48244388ce9 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -5828,6 +5828,7 @@ static inline int select_idle_smt(struct task_struct *p, int target)
  */
 static int select_idle_cpu(struct task_struct *p, struct sched_domain *sd, int target)
 {
+	struct cpumask *cpus = this_cpu_cpumask_var_ptr(select_idle_mask);
 	struct sched_domain *this_sd;
 	u64 avg_cost, avg_idle;
 	u64 time, cost;
@@ -5859,11 +5860,11 @@ static int select_idle_cpu(struct task_struct *p, struct sched_domain *sd, int t
 
 	time = cpu_clock(this);
 
-	for_each_cpu_wrap(cpu, sched_domain_span(sd), target) {
+	cpumask_and(cpus, sched_domain_span(sd), p->cpus_ptr);
+
+	for_each_cpu_wrap(cpu, cpus, target) {
 		if (!--nr)
 			return si_cpu;
-		if (!cpumask_test_cpu(cpu, p->cpus_ptr))
-			continue;
 		if (available_idle_cpu(cpu))
 			break;
 		if (si_cpu == -1 && sched_idle_cpu(cpu))
-- 
2.20.1

Re: [PATCH v2] sched/fair: Optimize select_idle_cpu

[Srikar Dronamraju]

* Cheng Jian <cj.chengjian@huawei.com> [2019-12-13 10:45:30]:

> Fixes: 1ad3aaf3fcd2 ("sched/core: Implement new approach to scale select_idle_cpu()")
> Signed-off-by: Cheng Jian <cj.chengjian@huawei.com>
> ---
>  kernel/sched/fair.c | 7 ++++---
>  1 file changed, 4 insertions(+), 3 deletions(-)
> 
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index 08a233e97a01..d48244388ce9 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -5828,6 +5828,7 @@ static inline int select_idle_smt(struct task_struct *p, int target)
>   */
>  static int select_idle_cpu(struct task_struct *p, struct sched_domain *sd, int target)
>  {
> +	struct cpumask *cpus = this_cpu_cpumask_var_ptr(select_idle_mask);
>  	struct sched_domain *this_sd;
>  	u64 avg_cost, avg_idle;
>  	u64 time, cost;
> @@ -5859,11 +5860,11 @@ static int select_idle_cpu(struct task_struct *p, struct sched_domain *sd, int t
>  
>  	time = cpu_clock(this);
>  
> -	for_each_cpu_wrap(cpu, sched_domain_span(sd), target) {
> +	cpumask_and(cpus, sched_domain_span(sd), p->cpus_ptr);
> +
> +	for_each_cpu_wrap(cpu, cpus, target) {
>  		if (!--nr)
>  			return si_cpu;
> -		if (!cpumask_test_cpu(cpu, p->cpus_ptr))
> -			continue;
>  		if (available_idle_cpu(cpu))
>  			break;
>  		if (si_cpu == -1 && sched_idle_cpu(cpu))
Looks good to me.

Reviewed-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>

-- 
Thanks and Regards
Srikar Dronamraju

Re: [PATCH v2] sched/fair: Optimize select_idle_cpu

[Vincent Guittot]

On Fri, 13 Dec 2019 at 03:48, Cheng Jian <cj.chengjian@huawei.com> wrote:
>
> select_idle_cpu() will scan the LLC domain for idle CPUs,
> it's always expensive. so the next commit :
>
>         1ad3aaf3fcd2 ("sched/core: Implement new approach to scale select_idle_cpu()")
>
> introduces a way to limit how many CPUs we scan.
>
> But it consume some CPUs out of 'nr' that are not allowed
> for the task and thus waste our attempts. The function
> always return nr_cpumask_bits, and we can't find a CPU
> which our task is allowed to run.
>
> Cpumask may be too big, similar to select_idle_core(), use
> per_cpu_ptr 'select_idle_mask' to prevent stack overflow.
>
> Fixes: 1ad3aaf3fcd2 ("sched/core: Implement new approach to scale select_idle_cpu()")
> Signed-off-by: Cheng Jian <cj.chengjian@huawei.com>

Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>

> ---
>  kernel/sched/fair.c | 7 ++++---
>  1 file changed, 4 insertions(+), 3 deletions(-)
>
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index 08a233e97a01..d48244388ce9 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -5828,6 +5828,7 @@ static inline int select_idle_smt(struct task_struct *p, int target)
>   */
>  static int select_idle_cpu(struct task_struct *p, struct sched_domain *sd, int target)
>  {
> +       struct cpumask *cpus = this_cpu_cpumask_var_ptr(select_idle_mask);
>         struct sched_domain *this_sd;
>         u64 avg_cost, avg_idle;
>         u64 time, cost;
> @@ -5859,11 +5860,11 @@ static int select_idle_cpu(struct task_struct *p, struct sched_domain *sd, int t
>
>         time = cpu_clock(this);
>
> -       for_each_cpu_wrap(cpu, sched_domain_span(sd), target) {
> +       cpumask_and(cpus, sched_domain_span(sd), p->cpus_ptr);
> +
> +       for_each_cpu_wrap(cpu, cpus, target) {
>                 if (!--nr)
>                         return si_cpu;
> -               if (!cpumask_test_cpu(cpu, p->cpus_ptr))
> -                       continue;
>                 if (available_idle_cpu(cpu))
>                         break;
>                 if (si_cpu == -1 && sched_idle_cpu(cpu))
> --
> 2.20.1
>

Re: [PATCH v2] sched/fair: Optimize select_idle_cpu

[Valentin Schneider]

On 13/12/2019 02:45, Cheng Jian wrote:
> select_idle_cpu() will scan the LLC domain for idle CPUs,
> it's always expensive. so the next commit :
> 
> 	1ad3aaf3fcd2 ("sched/core: Implement new approach to scale select_idle_cpu()")
> 
> introduces a way to limit how many CPUs we scan.
> 
> But it consume some CPUs out of 'nr' that are not allowed
> for the task and thus waste our attempts. The function
> always return nr_cpumask_bits, and we can't find a CPU
> which our task is allowed to run.
> 
> Cpumask may be too big, similar to select_idle_core(), use
> per_cpu_ptr 'select_idle_mask' to prevent stack overflow.
> 
> Fixes: 1ad3aaf3fcd2 ("sched/core: Implement new approach to scale select_idle_cpu()")
> Signed-off-by: Cheng Jian <cj.chengjian@huawei.com>

Reviewed-by: Valentin Schneider <valentin.schneider@arm.com>

> ---
>  kernel/sched/fair.c | 7 ++++---
>  1 file changed, 4 insertions(+), 3 deletions(-)
> 
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index 08a233e97a01..d48244388ce9 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -5828,6 +5828,7 @@ static inline int select_idle_smt(struct task_struct *p, int target)
>   */
>  static int select_idle_cpu(struct task_struct *p, struct sched_domain *sd, int target)
>  {
> +	struct cpumask *cpus = this_cpu_cpumask_var_ptr(select_idle_mask);
>  	struct sched_domain *this_sd;
>  	u64 avg_cost, avg_idle;
>  	u64 time, cost;
> @@ -5859,11 +5860,11 @@ static int select_idle_cpu(struct task_struct *p, struct sched_domain *sd, int t
>  
>  	time = cpu_clock(this);
>  
> -	for_each_cpu_wrap(cpu, sched_domain_span(sd), target) {
> +	cpumask_and(cpus, sched_domain_span(sd), p->cpus_ptr);
> +
> +	for_each_cpu_wrap(cpu, cpus, target) {
>  		if (!--nr)
>  			return si_cpu;
> -		if (!cpumask_test_cpu(cpu, p->cpus_ptr))
> -			continue;
>  		if (available_idle_cpu(cpu))
>  			break;
>  		if (si_cpu == -1 && sched_idle_cpu(cpu))
> 

[tip: sched/core] sched/fair: Optimize select_idle_cpu

[tip-bot2 for Cheng Jian]

The following commit has been merged into the sched/core branch of tip:

Commit-ID:     60588bfa223ff675b95f866249f90616613fbe31
Gitweb:        https://git.kernel.org/tip/60588bfa223ff675b95f866249f90616613fbe31
Author:        Cheng Jian <cj.chengjian@huawei.com>
AuthorDate:    Fri, 13 Dec 2019 10:45:30 +08:00
Committer:     Peter Zijlstra <peterz@infradead.org>
CommitterDate: Tue, 17 Dec 2019 13:32:51 +01:00

sched/fair: Optimize select_idle_cpu

select_idle_cpu() will scan the LLC domain for idle CPUs,
it's always expensive. so the next commit :

	1ad3aaf3fcd2 ("sched/core: Implement new approach to scale select_idle_cpu()")

introduces a way to limit how many CPUs we scan.

But it consume some CPUs out of 'nr' that are not allowed
for the task and thus waste our attempts. The function
always return nr_cpumask_bits, and we can't find a CPU
which our task is allowed to run.

Cpumask may be too big, similar to select_idle_core(), use
per_cpu_ptr 'select_idle_mask' to prevent stack overflow.

Fixes: 1ad3aaf3fcd2 ("sched/core: Implement new approach to scale select_idle_cpu()")
Signed-off-by: Cheng Jian <cj.chengjian@huawei.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Reviewed-by: Vincent Guittot <vincent.guittot@linaro.org>
Reviewed-by: Valentin Schneider <valentin.schneider@arm.com>
Link: https://lkml.kernel.org/r/20191213024530.28052-1-cj.chengjian@huawei.com
---
 kernel/sched/fair.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 846f50b..280d54c 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -5828,6 +5828,7 @@ static inline int select_idle_smt(struct task_struct *p, int target)
  */
 static int select_idle_cpu(struct task_struct *p, struct sched_domain *sd, int target)
 {
+	struct cpumask *cpus = this_cpu_cpumask_var_ptr(select_idle_mask);
 	struct sched_domain *this_sd;
 	u64 avg_cost, avg_idle;
 	u64 time, cost;
@@ -5859,11 +5860,11 @@ static int select_idle_cpu(struct task_struct *p, struct sched_domain *sd, int t
 
 	time = cpu_clock(this);
 
-	for_each_cpu_wrap(cpu, sched_domain_span(sd), target) {
+	cpumask_and(cpus, sched_domain_span(sd), p->cpus_ptr);
+
+	for_each_cpu_wrap(cpu, cpus, target) {
 		if (!--nr)
 			return si_cpu;
-		if (!cpumask_test_cpu(cpu, p->cpus_ptr))
-			continue;
 		if (available_idle_cpu(cpu))
 			break;
 		if (si_cpu == -1 && sched_idle_cpu(cpu))