[RFC] sched: Distinguish between idle_cpu calls based on desired effect

[RFC] sched: Distinguish between idle_cpu calls based on desired effect

[Rohit Jain]

In commit 247f2f6f3c706b40b5f3886646f3eb53671258bf we distinguish
between idle_cpu when the vcpu is not running for scheduling threads.
However, idle_cpu function is used at other places for actually checking
whether the state of the CPU is idle or not. Hence splitting the use of
that function based on the desired return value.

Signed-off-by: Rohit Jain <rohit.k.jain@oracle.com>
---
 include/linux/sched.h |  1 +
 kernel/sched/core.c   | 14 ++++++++++++++
 kernel/sched/fair.c   | 20 ++++++++++----------
 3 files changed, 25 insertions(+), 10 deletions(-)

diff --git a/include/linux/sched.h b/include/linux/sched.h
index c241370..959a858 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1504,6 +1504,7 @@ static inline int task_nice(const struct task_struct *p)
 extern int can_nice(const struct task_struct *p, const int nice);
 extern int task_curr(const struct task_struct *p);
 extern int idle_cpu(int cpu);
+extern int available_idle_cpu(int cpu);
 extern int sched_setscheduler(struct task_struct *, int, const struct sched_param *);
 extern int sched_setscheduler_nocheck(struct task_struct *, int, const struct sched_param *);
 extern int sched_setattr(struct task_struct *, const struct sched_attr *);
diff --git a/kernel/sched/core.c b/kernel/sched/core.c
index 71bdb86..171e160 100644
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -4028,6 +4028,20 @@ int idle_cpu(int cpu)
 		return 0;
 #endif
 
+	return 1;
+}
+
+/**
+ * idle_cpu - is a given CPU idle for enqueuing work.
+ * @cpu: the processor in question.
+ *
+ * Return: 1 if the CPU is currently idle. 0 otherwise.
+ */
+int available_idle_cpu(int cpu)
+{
+	if (!idle_cpu(cpu))
+		return 0;
+
 	if (vcpu_is_preempted(cpu))
 		return 0;
 
diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
index 1f6a23a..a0b0da4 100644
--- a/kernel/sched/fair.c
+++ b/kernel/sched/fair.c
@@ -5870,8 +5870,8 @@ wake_affine_idle(int this_cpu, int prev_cpu, int sync)
 	 * a cpufreq perspective, it's better to have higher utilisation
 	 * on one CPU.
 	 */
-	if (idle_cpu(this_cpu) && cpus_share_cache(this_cpu, prev_cpu))
-		return idle_cpu(prev_cpu) ? prev_cpu : this_cpu;
+	if (available_idle_cpu(this_cpu) && cpus_share_cache(this_cpu, prev_cpu))
+		return available_idle_cpu(prev_cpu) ? prev_cpu : this_cpu;
 
 	if (sync && cpu_rq(this_cpu)->nr_running == 1)
 		return this_cpu;
@@ -6157,7 +6157,7 @@ find_idlest_group_cpu(struct sched_group *group, struct task_struct *p, int this
 
 	/* Traverse only the allowed CPUs */
 	for_each_cpu_and(i, sched_group_span(group), &p->cpus_allowed) {
-		if (idle_cpu(i)) {
+		if (available_idle_cpu(i)) {
 			struct rq *rq = cpu_rq(i);
 			struct cpuidle_state *idle = idle_get_state(rq);
 			if (idle && idle->exit_latency < min_exit_latency) {
@@ -6286,7 +6286,7 @@ void __update_idle_core(struct rq *rq)
 		if (cpu == core)
 			continue;
 
-		if (!idle_cpu(cpu))
+		if (!available_idle_cpu(cpu))
 			goto unlock;
 	}
 
@@ -6318,7 +6318,7 @@ static int select_idle_core(struct task_struct *p, struct sched_domain *sd, int
 
 		for_each_cpu(cpu, cpu_smt_mask(core)) {
 			cpumask_clear_cpu(cpu, cpus);
-			if (!idle_cpu(cpu))
+			if (!available_idle_cpu(cpu))
 				idle = false;
 		}
 
@@ -6347,7 +6347,7 @@ static int select_idle_smt(struct task_struct *p, struct sched_domain *sd, int t
 	for_each_cpu(cpu, cpu_smt_mask(target)) {
 		if (!cpumask_test_cpu(cpu, &p->cpus_allowed))
 			continue;
-		if (idle_cpu(cpu))
+		if (available_idle_cpu(cpu))
 			return cpu;
 	}
 
@@ -6410,7 +6410,7 @@ static int select_idle_cpu(struct task_struct *p, struct sched_domain *sd, int t
 			return -1;
 		if (!cpumask_test_cpu(cpu, &p->cpus_allowed))
 			continue;
-		if (idle_cpu(cpu))
+		if (available_idle_cpu(cpu))
 			break;
 	}
 
@@ -6430,13 +6430,13 @@ static int select_idle_sibling(struct task_struct *p, int prev, int target)
 	struct sched_domain *sd;
 	int i, recent_used_cpu;
 
-	if (idle_cpu(target))
+	if (available_idle_cpu(target))
 		return target;
 
 	/*
 	 * If the previous CPU is cache affine and idle, don't be stupid:
 	 */
-	if (prev != target && cpus_share_cache(prev, target) && idle_cpu(prev))
+	if (prev != target && cpus_share_cache(prev, target) && available_idle_cpu(prev))
 		return prev;
 
 	/* Check a recently used CPU as a potential idle candidate: */
@@ -6444,7 +6444,7 @@ static int select_idle_sibling(struct task_struct *p, int prev, int target)
 	if (recent_used_cpu != prev &&
 	    recent_used_cpu != target &&
 	    cpus_share_cache(recent_used_cpu, target) &&
-	    idle_cpu(recent_used_cpu) &&
+	    available_idle_cpu(recent_used_cpu) &&
 	    cpumask_test_cpu(p->recent_used_cpu, &p->cpus_allowed)) {
 		/*
 		 * Replace recent_used_cpu with prev as it is a potential
-- 
2.7.4

Re: [RFC] sched: Distinguish between idle_cpu calls based on desired effect

[Randy Dunlap]

On 05/07/2018 04:41 PM, Rohit Jain wrote:
> In commit 247f2f6f3c706b40b5f3886646f3eb53671258bf we distinguish
> between idle_cpu when the vcpu is not running for scheduling threads.
> However, idle_cpu function is used at other places for actually checking
> whether the state of the CPU is idle or not. Hence splitting the use of
> that function based on the desired return value.
> 
> Signed-off-by: Rohit Jain <rohit.k.jain@oracle.com>
> ---
>  include/linux/sched.h |  1 +
>  kernel/sched/core.c   | 14 ++++++++++++++
>  kernel/sched/fair.c   | 20 ++++++++++----------
>  3 files changed, 25 insertions(+), 10 deletions(-)
> 

> diff --git a/kernel/sched/core.c b/kernel/sched/core.c
> index 71bdb86..171e160 100644
> --- a/kernel/sched/core.c
> +++ b/kernel/sched/core.c
> @@ -4028,6 +4028,20 @@ int idle_cpu(int cpu)
>  		return 0;
>  #endif
>  
> +	return 1;
> +}
> +
> +/**
> + * idle_cpu - is a given CPU idle for enqueuing work.

    * available_idle_cpu -

> + * @cpu: the processor in question.
> + *
> + * Return: 1 if the CPU is currently idle. 0 otherwise.
> + */
> +int available_idle_cpu(int cpu)
> +{
> +	if (!idle_cpu(cpu))
> +		return 0;
> +
>  	if (vcpu_is_preempted(cpu))
>  		return 0;
>  


-- 
~Randy

Re: [RFC] sched: Distinguish between idle_cpu calls based on desired effect

[Rohit Jain]

Hi Randy,


On 05/07/2018 05:15 PM, Randy Dunlap wrote:
> On 05/07/2018 04:41 PM, Rohit Jain wrote:

<snip>

>> diff --git a/kernel/sched/core.c b/kernel/sched/core.c
>> index 71bdb86..171e160 100644
>> --- a/kernel/sched/core.c
>> +++ b/kernel/sched/core.c
>> @@ -4028,6 +4028,20 @@ int idle_cpu(int cpu)
>>   		return 0;
>>   #endif
>>   
>> +	return 1;
>> +}
>> +
>> +/**
>> + * idle_cpu - is a given CPU idle for enqueuing work.
>      * available_idle_cpu -

Thanks for the catch!

Will fix it in the next version.

Thanks,
Rohit

<snip>

Re: [RFC] sched: Distinguish between idle_cpu calls based on desired effect

[Peter Zijlstra]

On Mon, May 07, 2018 at 04:41:17PM -0700, Rohit Jain wrote:
> In commit 247f2f6f3c706b40b5f3886646f3eb53671258bf we distinguish
> between idle_cpu when the vcpu is not running for scheduling threads.
> However, idle_cpu function is used at other places for actually checking
> whether the state of the CPU is idle or not. Hence splitting the use of
> that function based on the desired return value.

Yes, that looks OK.

for your gitconfig:

[core]
	abbrev = 12
[alias]
        one = show -s --pretty='format:%h (\"%s\")'

$ git one 247f2f6f3c706b40b5f3886646f3eb53671258bf
247f2f6f3c70 ("sched/core: Don't schedule threads on pre-empted vCPUs")

which gives the normal commit quote style.

Re: [RFC] sched: Distinguish between idle_cpu calls based on desired effect

[Rohit Jain]

Hi Peter,


On 05/09/2018 01:51 AM, Peter Zijlstra wrote:
> On Mon, May 07, 2018 at 04:41:17PM -0700, Rohit Jain wrote:
>> In commit 247f2f6f3c706b40b5f3886646f3eb53671258bf we distinguish
>> between idle_cpu when the vcpu is not running for scheduling threads.
>> However, idle_cpu function is used at other places for actually checking
>> whether the state of the CPU is idle or not. Hence splitting the use of
>> that function based on the desired return value.
> Yes, that looks OK.

Great! Let me change the comment and send out v2.

>
> for your gitconfig:
>
> [core]
> 	abbrev = 12
> [alias]
>          one = show -s --pretty='format:%h (\"%s\")'
>
> $ git one 247f2f6f3c706b40b5f3886646f3eb53671258bf
> 247f2f6f3c70 ("sched/core: Don't schedule threads on pre-empted vCPUs")
>
> which gives the normal commit quote style.
>

Thanks! updated :)

Regards,
Rohit