[RESEND PATCH v2 1/2] sched/deadline: Add cpudl_maximum_dl() for clean-up

[RESEND PATCH v2 1/2] sched/deadline: Add cpudl_maximum_dl() for clean-up

[Byungchul Park]

Changes from v1
 - Enhance commit msg
 - Prevent WARN in cpumask_test_cpu() in cpudl_find() when best_cpu == -1

-----8<-----
>From 7735382d07ae6a61d740ae39ba2ecf169d43b8a2 Mon Sep 17 00:00:00 2001
From: Byungchul Park <byungchul.park@lge.com>
Date: Wed, 22 Mar 2017 14:25:56 +0900
Subject: [RESEND PATCH v2 1/2] sched/deadline: Add cpudl_maximum_dl() for clean-up

Current code uses cpudl_maximum() to get the root node's cpu, while it
directly accesses the root node like 'cp->elements[0].dl' to get the
root node's dl. It would be more readible to add a function for the dl,
as well. Added it.

Signed-off-by: Byungchul Park <byungchul.park@lge.com>
Acked-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
---
 kernel/sched/cpudeadline.c | 11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

diff --git a/kernel/sched/cpudeadline.c b/kernel/sched/cpudeadline.c
index 8d9562d..9f02035 100644
--- a/kernel/sched/cpudeadline.c
+++ b/kernel/sched/cpudeadline.c
@@ -108,11 +108,16 @@ static void cpudl_heapify(struct cpudl *cp, int idx)
 		cpudl_heapify_down(cp, idx);
 }
 
-static inline int cpudl_maximum(struct cpudl *cp)
+static inline int cpudl_maximum_cpu(struct cpudl *cp)
 {
 	return cp->elements[0].cpu;
 }
 
+static inline u64 cpudl_maximum_dl(struct cpudl *cp)
+{
+	return cp->elements[0].dl;
+}
+
 /*
  * cpudl_find - find the best (later-dl) CPU in the system
  * @cp: the cpudl max-heap context
@@ -130,11 +135,11 @@ int cpudl_find(struct cpudl *cp, struct task_struct *p,
 	    cpumask_and(later_mask, cp->free_cpus, &p->cpus_allowed)) {
 		return 1;
 	} else {
-		int best_cpu = cpudl_maximum(cp);
+		int best_cpu = cpudl_maximum_cpu(cp);
 		WARN_ON(best_cpu != -1 && !cpu_present(best_cpu));
 
 		if (cpumask_test_cpu(best_cpu, &p->cpus_allowed) &&
-		    dl_time_before(dl_se->deadline, cp->elements[0].dl)) {
+		    dl_time_before(dl_se->deadline, cpudl_maximum_dl(cp))) {
 			if (later_mask)
 				cpumask_set_cpu(best_cpu, later_mask);
 
-- 
1.9.1

[RESEND PATCH v2 2/2] sched/deadline: Initialize cp->elements[].cpu to an invalid value

[Byungchul Park]

Currently, migrating tasks to cpu0 unconditionally happens when the
heap is empty, since cp->elements[].cpu was initialized to 0(=cpu0).
We have to distinguish between the empty case and cpu0 to avoid the
unnecessary migrantions. Therefore, it has to return an invalid value
e.i. -1 in that case.

Signed-off-by: Byungchul Park <byungchul.park@lge.com>
Acked-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
---
 kernel/sched/cpudeadline.c | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/kernel/sched/cpudeadline.c b/kernel/sched/cpudeadline.c
index 9f02035..bcf903f 100644
--- a/kernel/sched/cpudeadline.c
+++ b/kernel/sched/cpudeadline.c
@@ -138,6 +138,12 @@ int cpudl_find(struct cpudl *cp, struct task_struct *p,
 		int best_cpu = cpudl_maximum_cpu(cp);
 		WARN_ON(best_cpu != -1 && !cpu_present(best_cpu));
 
+		/*
+		 * The heap tree is empry for now, just return.
+		 */
+		if (best_cpu == -1)
+			return 0;
+
 		if (cpumask_test_cpu(best_cpu, &p->cpus_allowed) &&
 		    dl_time_before(dl_se->deadline, cpudl_maximum_dl(cp))) {
 			if (later_mask)
@@ -265,8 +271,10 @@ int cpudl_init(struct cpudl *cp)
 		return -ENOMEM;
 	}
 
-	for_each_possible_cpu(i)
+	for_each_possible_cpu(i) {
+		cp->elements[i].cpu = -1;
 		cp->elements[i].idx = IDX_INVALID;
+	}
 
 	return 0;
 }
-- 
1.9.1

Re: [RESEND PATCH v2 2/2] sched/deadline: Initialize cp->elements[].cpu to an invalid value

[Byungchul Park]

On Tue, Oct 10, 2017 at 03:44:04PM +0900, Byungchul Park wrote:
> Currently, migrating tasks to cpu0 unconditionally happens when the
> heap is empty, since cp->elements[].cpu was initialized to 0(=cpu0).
> We have to distinguish between the empty case and cpu0 to avoid the
> unnecessary migrantions. Therefore, it has to return an invalid value
> e.i. -1 in that case.

This patch does what it should obviously do. Could you check this?

> Signed-off-by: Byungchul Park <byungchul.park@lge.com>
> Acked-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
> ---
>  kernel/sched/cpudeadline.c | 10 +++++++++-
>  1 file changed, 9 insertions(+), 1 deletion(-)
> 
> diff --git a/kernel/sched/cpudeadline.c b/kernel/sched/cpudeadline.c
> index 9f02035..bcf903f 100644
> --- a/kernel/sched/cpudeadline.c
> +++ b/kernel/sched/cpudeadline.c
> @@ -138,6 +138,12 @@ int cpudl_find(struct cpudl *cp, struct task_struct *p,
>  		int best_cpu = cpudl_maximum_cpu(cp);
>  		WARN_ON(best_cpu != -1 && !cpu_present(best_cpu));
>  
> +		/*
> +		 * The heap tree is empry for now, just return.
> +		 */
> +		if (best_cpu == -1)
> +			return 0;
> +
>  		if (cpumask_test_cpu(best_cpu, &p->cpus_allowed) &&
>  		    dl_time_before(dl_se->deadline, cpudl_maximum_dl(cp))) {
>  			if (later_mask)
> @@ -265,8 +271,10 @@ int cpudl_init(struct cpudl *cp)
>  		return -ENOMEM;
>  	}
>  
> -	for_each_possible_cpu(i)
> +	for_each_possible_cpu(i) {
> +		cp->elements[i].cpu = -1;
>  		cp->elements[i].idx = IDX_INVALID;
> +	}
>  
>  	return 0;
>  }
> -- 
> 1.9.1

Re: [RESEND PATCH v2 1/2] sched/deadline: Add cpudl_maximum_dl() for clean-up

[Byungchul Park]

On Tue, Oct 10, 2017 at 03:44:03PM +0900, Byungchul Park wrote:
> Changes from v1
>  - Enhance commit msg
>  - Prevent WARN in cpumask_test_cpu() in cpudl_find() when best_cpu == -1
> 
> -----8<-----
> >From 7735382d07ae6a61d740ae39ba2ecf169d43b8a2 Mon Sep 17 00:00:00 2001
> From: Byungchul Park <byungchul.park@lge.com>
> Date: Wed, 22 Mar 2017 14:25:56 +0900
> Subject: [RESEND PATCH v2 1/2] sched/deadline: Add cpudl_maximum_dl() for clean-up
> 
> Current code uses cpudl_maximum() to get the root node's cpu, while it
> directly accesses the root node like 'cp->elements[0].dl' to get the
> root node's dl. It would be more readible to add a function for the dl,
> as well. Added it.

Is there any opinion about this patch? If no, could you consider this?

> 
> Signed-off-by: Byungchul Park <byungchul.park@lge.com>
> Acked-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
> ---
>  kernel/sched/cpudeadline.c | 11 ++++++++---
>  1 file changed, 8 insertions(+), 3 deletions(-)
> 
> diff --git a/kernel/sched/cpudeadline.c b/kernel/sched/cpudeadline.c
> index 8d9562d..9f02035 100644
> --- a/kernel/sched/cpudeadline.c
> +++ b/kernel/sched/cpudeadline.c
> @@ -108,11 +108,16 @@ static void cpudl_heapify(struct cpudl *cp, int idx)
>  		cpudl_heapify_down(cp, idx);
>  }
>  
> -static inline int cpudl_maximum(struct cpudl *cp)
> +static inline int cpudl_maximum_cpu(struct cpudl *cp)
>  {
>  	return cp->elements[0].cpu;
>  }
>  
> +static inline u64 cpudl_maximum_dl(struct cpudl *cp)
> +{
> +	return cp->elements[0].dl;
> +}
> +
>  /*
>   * cpudl_find - find the best (later-dl) CPU in the system
>   * @cp: the cpudl max-heap context
> @@ -130,11 +135,11 @@ int cpudl_find(struct cpudl *cp, struct task_struct *p,
>  	    cpumask_and(later_mask, cp->free_cpus, &p->cpus_allowed)) {
>  		return 1;
>  	} else {
> -		int best_cpu = cpudl_maximum(cp);
> +		int best_cpu = cpudl_maximum_cpu(cp);
>  		WARN_ON(best_cpu != -1 && !cpu_present(best_cpu));
>  
>  		if (cpumask_test_cpu(best_cpu, &p->cpus_allowed) &&
> -		    dl_time_before(dl_se->deadline, cp->elements[0].dl)) {
> +		    dl_time_before(dl_se->deadline, cpudl_maximum_dl(cp))) {
>  			if (later_mask)
>  				cpumask_set_cpu(best_cpu, later_mask);
>  
> -- 
> 1.9.1

Re: [RESEND PATCH v2 1/2] sched/deadline: Add cpudl_maximum_dl() for clean-up

[Byungchul Park]

On Tue, Oct 10, 2017 at 03:44:03PM +0900, Byungchul Park wrote:
> Changes from v1
>  - Enhance commit msg
>  - Prevent WARN in cpumask_test_cpu() in cpudl_find() when best_cpu == -1
> 
> -----8<-----
> >From 7735382d07ae6a61d740ae39ba2ecf169d43b8a2 Mon Sep 17 00:00:00 2001
> From: Byungchul Park <byungchul.park@lge.com>
> Date: Wed, 22 Mar 2017 14:25:56 +0900
> Subject: [RESEND PATCH v2 1/2] sched/deadline: Add cpudl_maximum_dl() for clean-up
> 
> Current code uses cpudl_maximum() to get the root node's cpu, while it
> directly accesses the root node like 'cp->elements[0].dl' to get the
> root node's dl. It would be more readible to add a function for the dl,
> as well. Added it.

Hello everyone,

Could you tell opinions?

> Signed-off-by: Byungchul Park <byungchul.park@lge.com>
> Acked-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
> ---
>  kernel/sched/cpudeadline.c | 11 ++++++++---
>  1 file changed, 8 insertions(+), 3 deletions(-)
> 
> diff --git a/kernel/sched/cpudeadline.c b/kernel/sched/cpudeadline.c
> index 8d9562d..9f02035 100644
> --- a/kernel/sched/cpudeadline.c
> +++ b/kernel/sched/cpudeadline.c
> @@ -108,11 +108,16 @@ static void cpudl_heapify(struct cpudl *cp, int idx)
>  		cpudl_heapify_down(cp, idx);
>  }
>  
> -static inline int cpudl_maximum(struct cpudl *cp)
> +static inline int cpudl_maximum_cpu(struct cpudl *cp)
>  {
>  	return cp->elements[0].cpu;
>  }
>  
> +static inline u64 cpudl_maximum_dl(struct cpudl *cp)
> +{
> +	return cp->elements[0].dl;
> +}
> +
>  /*
>   * cpudl_find - find the best (later-dl) CPU in the system
>   * @cp: the cpudl max-heap context
> @@ -130,11 +135,11 @@ int cpudl_find(struct cpudl *cp, struct task_struct *p,
>  	    cpumask_and(later_mask, cp->free_cpus, &p->cpus_allowed)) {
>  		return 1;
>  	} else {
> -		int best_cpu = cpudl_maximum(cp);
> +		int best_cpu = cpudl_maximum_cpu(cp);
>  		WARN_ON(best_cpu != -1 && !cpu_present(best_cpu));
>  
>  		if (cpumask_test_cpu(best_cpu, &p->cpus_allowed) &&
> -		    dl_time_before(dl_se->deadline, cp->elements[0].dl)) {
> +		    dl_time_before(dl_se->deadline, cpudl_maximum_dl(cp))) {
>  			if (later_mask)
>  				cpumask_set_cpu(best_cpu, later_mask);
>  
> -- 
> 1.9.1

Re: [RESEND PATCH v2 1/2] sched/deadline: Add cpudl_maximum_dl() for clean-up

[Daniel Bristot de Oliveira]

On 10/27/2017 10:38 AM, Byungchul Park wrote:
> On Tue, Oct 10, 2017 at 03:44:03PM +0900, Byungchul Park wrote:
>> Changes from v1
>>  - Enhance commit msg
>>  - Prevent WARN in cpumask_test_cpu() in cpudl_find() when best_cpu == -1
>>
>> -----8<-----
>> >From 7735382d07ae6a61d740ae39ba2ecf169d43b8a2 Mon Sep 17 00:00:00 2001
>> From: Byungchul Park <byungchul.park@lge.com>
>> Date: Wed, 22 Mar 2017 14:25:56 +0900
>> Subject: [RESEND PATCH v2 1/2] sched/deadline: Add cpudl_maximum_dl() for clean-up
>>
>> Current code uses cpudl_maximum() to get the root node's cpu, while it
>> directly accesses the root node like 'cp->elements[0].dl' to get the
>> root node's dl. It would be more readible to add a function for the dl,
>> as well. Added it.
> 
> Hello everyone,
> 
> Could you tell opinions?
> 
>> Signed-off-by: Byungchul Park <byungchul.park@lge.com>
>> Acked-by: Steven Rostedt (VMware) <rostedt@goodmis.org>

Acked-by: Daniel Bristot de Oliveira <bristot@redhat.com>

>> ---
>>  kernel/sched/cpudeadline.c | 11 ++++++++---
>>  1 file changed, 8 insertions(+), 3 deletions(-)
>>
>> diff --git a/kernel/sched/cpudeadline.c b/kernel/sched/cpudeadline.c
>> index 8d9562d..9f02035 100644
>> --- a/kernel/sched/cpudeadline.c
>> +++ b/kernel/sched/cpudeadline.c
>> @@ -108,11 +108,16 @@ static void cpudl_heapify(struct cpudl *cp, int idx)
>>  		cpudl_heapify_down(cp, idx);
>>  }
>>  
>> -static inline int cpudl_maximum(struct cpudl *cp)
>> +static inline int cpudl_maximum_cpu(struct cpudl *cp)
>>  {
>>  	return cp->elements[0].cpu;
>>  }
>>  
>> +static inline u64 cpudl_maximum_dl(struct cpudl *cp)
>> +{
>> +	return cp->elements[0].dl;
>> +}
>> +
>>  /*
>>   * cpudl_find - find the best (later-dl) CPU in the system
>>   * @cp: the cpudl max-heap context
>> @@ -130,11 +135,11 @@ int cpudl_find(struct cpudl *cp, struct task_struct *p,
>>  	    cpumask_and(later_mask, cp->free_cpus, &p->cpus_allowed)) {
>>  		return 1;
>>  	} else {
>> -		int best_cpu = cpudl_maximum(cp);
>> +		int best_cpu = cpudl_maximum_cpu(cp);
>>  		WARN_ON(best_cpu != -1 && !cpu_present(best_cpu));
>>  
>>  		if (cpumask_test_cpu(best_cpu, &p->cpus_allowed) &&
>> -		    dl_time_before(dl_se->deadline, cp->elements[0].dl)) {
>> +		    dl_time_before(dl_se->deadline, cpudl_maximum_dl(cp))) {
>>  			if (later_mask)
>>  				cpumask_set_cpu(best_cpu, later_mask);
>>  
>> -- 
>> 1.9.1

Re: [RESEND PATCH v2 2/2] sched/deadline: Initialize cp->elements[].cpu to an invalid value

[Daniel Bristot de Oliveira]

On 10/12/2017 09:57 AM, Byungchul Park wrote:
> On Tue, Oct 10, 2017 at 03:44:04PM +0900, Byungchul Park wrote:
>> Currently, migrating tasks to cpu0 unconditionally happens when the
>> heap is empty, since cp->elements[].cpu was initialized to 0(=cpu0).
>> We have to distinguish between the empty case and cpu0 to avoid the
>> unnecessary migrantions. Therefore, it has to return an invalid value

   s/migrantions/migrations/ ?

>> e.i. -1 in that case.
> 
> This patch does what it should obviously do. Could you check this?
> 
>> Signed-off-by: Byungchul Park <byungchul.park@lge.com>
>> Acked-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
>> ---
>>  kernel/sched/cpudeadline.c | 10 +++++++++-
>>  1 file changed, 9 insertions(+), 1 deletion(-)
>>
>> diff --git a/kernel/sched/cpudeadline.c b/kernel/sched/cpudeadline.c
>> index 9f02035..bcf903f 100644
>> --- a/kernel/sched/cpudeadline.c
>> +++ b/kernel/sched/cpudeadline.c
>> @@ -138,6 +138,12 @@ int cpudl_find(struct cpudl *cp, struct task_struct *p,
>>  		int best_cpu = cpudl_maximum_cpu(cp);
>>  		WARN_ON(best_cpu != -1 && !cpu_present(best_cpu));
>>  
>> +		/*
>> +		 * The heap tree is empry for now, just return.

		  s/empry/empty/ ?

>> +		 */
>> +		if (best_cpu == -1)
>> +			return 0;
>> +
>>  		if (cpumask_test_cpu(best_cpu, &p->cpus_allowed) &&
>>  		    dl_time_before(dl_se->deadline, cpudl_maximum_dl(cp))) {
>>  			if (later_mask)
>> @@ -265,8 +271,10 @@ int cpudl_init(struct cpudl *cp)
>>  		return -ENOMEM;
>>  	}
>>  
>> -	for_each_possible_cpu(i)
>> +	for_each_possible_cpu(i) {
>> +		cp->elements[i].cpu = -1;
>>  		cp->elements[i].idx = IDX_INVALID;
>> +	}
>>  
>>  	return 0;
>>  }
>> -- 
>> 1.9.1

Re: [RESEND PATCH v2 2/2] sched/deadline: Initialize cp->elements[].cpu to an invalid value

[Byungchul Park]

On 11/13/2017 9:46 PM, Daniel Bristot de Oliveira wrote:
> On 10/12/2017 09:57 AM, Byungchul Park wrote:
>> On Tue, Oct 10, 2017 at 03:44:04PM +0900, Byungchul Park wrote:
>>> Currently, migrating tasks to cpu0 unconditionally happens when the
>>> heap is empty, since cp->elements[].cpu was initialized to 0(=cpu0).
>>> We have to distinguish between the empty case and cpu0 to avoid the
>>> unnecessary migrantions. Therefore, it has to return an invalid value
> 
>     s/migrantions/migrations/ ?
> 
>>> e.i. -1 in that case.
>>
>> This patch does what it should obviously do. Could you check this?
>>
>>> Signed-off-by: Byungchul Park <byungchul.park@lge.com>
>>> Acked-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
>>> ---
>>>   kernel/sched/cpudeadline.c | 10 +++++++++-
>>>   1 file changed, 9 insertions(+), 1 deletion(-)
>>>
>>> diff --git a/kernel/sched/cpudeadline.c b/kernel/sched/cpudeadline.c
>>> index 9f02035..bcf903f 100644
>>> --- a/kernel/sched/cpudeadline.c
>>> +++ b/kernel/sched/cpudeadline.c
>>> @@ -138,6 +138,12 @@ int cpudl_find(struct cpudl *cp, struct task_struct *p,
>>>   		int best_cpu = cpudl_maximum_cpu(cp);
>>>   		WARN_ON(best_cpu != -1 && !cpu_present(best_cpu));
>>>   
>>> +		/*
>>> +		 * The heap tree is empry for now, just return.
> 
> 		  s/empry/empty/ ?

Hello Daniel,

Thank you very much for pointing out typos.

I shall do spellcheck evetytime submiting patches from now on.

-- 
Thanks,
Byungchul

[RESEND PATCH v2 2/2] sched/deadline: Initialize cp->elements[].cpu to an invalid value

[Byungchul Park]

Currently, migrating tasks to cpu0 unconditionally happens when the
heap is empty, since cp->elements[].cpu was initialized to 0(=cpu0).
We have to distinguish between the empty case and cpu0 to avoid the
unnecessary migrantions. Therefore, it has to return an invalid value
e.i. -1 in that case.

Signed-off-by: Byungchul Park <byungchul.park@lge.com>
Acked-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
---
 kernel/sched/cpudeadline.c | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/kernel/sched/cpudeadline.c b/kernel/sched/cpudeadline.c
index 9f02035..bcf903f 100644
--- a/kernel/sched/cpudeadline.c
+++ b/kernel/sched/cpudeadline.c
@@ -138,6 +138,12 @@ int cpudl_find(struct cpudl *cp, struct task_struct *p,
 		int best_cpu = cpudl_maximum_cpu(cp);
 		WARN_ON(best_cpu != -1 && !cpu_present(best_cpu));
 
+		/*
+		 * The heap tree is empry for now, just return.
+		 */
+		if (best_cpu == -1)
+			return 0;
+
 		if (cpumask_test_cpu(best_cpu, &p->cpus_allowed) &&
 		    dl_time_before(dl_se->deadline, cpudl_maximum_dl(cp))) {
 			if (later_mask)
@@ -265,8 +271,10 @@ int cpudl_init(struct cpudl *cp)
 		return -ENOMEM;
 	}
 
-	for_each_possible_cpu(i)
+	for_each_possible_cpu(i) {
+		cp->elements[i].cpu = -1;
 		cp->elements[i].idx = IDX_INVALID;
+	}
 
 	return 0;
 }
-- 
1.9.1