Skip to contentSkip to conversation

sched/fair: Limit access to overutilized

PATCH Applied Shrikanth Hegde 5 participants 15 messages Mar 7, 2024 – Mar 26, 2024 Scheduler

lore
Share & tools
lore
#0 lore raw

v6 cover sched/fair: Limit access to overutilized

When running an ISV workload on a large system (240 Cores, SMT8), it was observed from perf profile that newidle_balance and enqueue_task_fair were consuming more cycles. Perf annotate showed that most of the time was spent on accessing overutilized field of root domain. Similar perf profile was simulated by making some changes to stress-ng --wait. Both newidle_balance and enqueue_task_fair consume close to 5-7%.

Overutilized was added for EAS(Energy aware scheduler) to decide whether to do load balance or not. Simultaneous access to overutilized by multiple CPUs lead cache invalidations due to true sharing. Updating overutilized is not required for non-EAS platforms. Since overutilized and overload are part of the same cacheline, there is false sharing as well.

Patch 1/3 - Main patch. It helps in reducing the above said issue. Both the functions don't show up in the profile. With patch comparison is in changelog. With the patch stated problem in the ISV workload also got solved and throughput has improved.
Patch 2/3 - Code refactoring to use the helper function instead of direct access of the field. Keeping this patch so patch 3/3 becomes easier to understand. Depends on 1/3.
Patch 3/3 - Refactoring the code since most of the patterns are observed are eas && !overutilzed. Changed the helper function accordingly. Depends on 2/3.

More details can be found in cover letter of v1.

v5 -> v6:
- Mades minor changes for !CONFIG_SMP cases as pointed out by Vincent. v4 -> v5:
- Added EAS check inside helper functions instead of sprinkling it around. EAS check is static branch at best. So that keeps the code simpler.
- added EAS check inside cpu_overutilized as suggested by Qais.
- Added patch 3 since most of the code does eas && !overutilized pattern as suggested by Vincent.
v3 -> v4:
- corrected a mistake where EAS check was missed.
v2 -> v3:
- Pierre and Dietmar suggested we could add one more EAS check before calling cpu_overutilized. That makes sense since that value is not used anyway in Non-EAS case.
- Refactored the code as dietmar suggested to avoid additional call to sched_energy_enabled().
- Minor edits to change log.
v1 -> v2:
Chen Yu pointed out minor issue in code. Corrected that code and updated the changelog.

v1: https://lore.kernel.org/lkml/[email protected]/
v2: https://lore.kernel.org/lkml/[email protected]/
v3: https://lore.kernel.org/lkml/[email protected]/
v4: https://lore.kernel.org/lkml/[email protected]/
v5: https://lore.kernel.org/lkml/[email protected]/

Shrikanth Hegde (3):
sched/fair: Add EAS checks before updating overutilized sched/fair: Use helper function to access rd->overutilized sched/fair: Combine EAS check with overutilized access

 kernel/sched/fair.c | 72 ++++++++++++++++++++++++++++-----------------
 1 file changed, 45 insertions(+), 27 deletions(-)

--
2.39.3

#1 lore raw

PATCH v6 1/3 sched/fair: Add EAS checks before updating overutilized

Overutilized field of root domain is only used for EAS(energy aware scheduler) to decide whether to do load balance or not. It is not used if EAS not possible.

Currently enqueue_task_fair and task_tick_fair accesses, sometime updates this field. In update_sd_lb_stats it is updated often. This causes cache contention due to true sharing and burns a lot of cycles. overload and overutilized are part of the same cacheline. Updating it often invalidates the cacheline. That causes access to overload to slow down due to false sharing. Hence add EAS check before accessing/updating this field. EAS check is optimized at compile time or it is a static branch. Hence it shouldn't cost much.

With the patch, both enqueue_task_fair and newidle_balance don't show up as hot routines in perf profile.

6.8-rc4:

7.18%  swapper          [kernel.vmlinux]              [k] enqueue_task_fair
6.78%  s                [kernel.vmlinux]              [k] newidle_balance
+patch:
0.14%  swapper          [kernel.vmlinux]              [k] enqueue_task_fair
0.00%  swapper          [kernel.vmlinux]              [k] newidle_balance

Minor change: trace_sched_overutilized_tp expect that second argument to be bool. So do a int to bool conversion for that.

Fixes 2802bf3cd936 ("sched/fair: Add over-utilization/tipping point indicator")
---
 kernel/sched/fair.c | 53 +++++++++++++++++++++++++++++----------------
 1 file changed, 34 insertions(+), 19 deletions(-)
kernel/sched/fair.c +34 -19
index 6a16129f9a5c..5aa6e918598c 100644
@@ -6663,22 +6663,42 @@ static inline void hrtick_update(struct rq *rq)
 #ifdef CONFIG_SMP
 static inline bool cpu_overutilized(int cpu)
 {
-	unsigned long rq_util_min = uclamp_rq_get(cpu_rq(cpu), UCLAMP_MIN);
-	unsigned long rq_util_max = uclamp_rq_get(cpu_rq(cpu), UCLAMP_MAX);
+	unsigned long  rq_util_min, rq_util_max;
+
+	if (!sched_energy_enabled())
+		return false;
+
+	rq_util_min = uclamp_rq_get(cpu_rq(cpu), UCLAMP_MIN);
+	rq_util_max = uclamp_rq_get(cpu_rq(cpu), UCLAMP_MAX);
 
 	/* Return true only if the utilization doesn't fit CPU's capacity */
 	return !util_fits_cpu(cpu_util_cfs(cpu), rq_util_min, rq_util_max, cpu);
 }
 
-static inline void update_overutilized_status(struct rq *rq)
+static inline void set_rd_overutilized_status(struct root_domain *rd,
+					      unsigned int status)
 {
-	if (!READ_ONCE(rq->rd->overutilized) && cpu_overutilized(rq->cpu)) {
-		WRITE_ONCE(rq->rd->overutilized, SG_OVERUTILIZED);
-		trace_sched_overutilized_tp(rq->rd, SG_OVERUTILIZED);
-	}
+	if (!sched_energy_enabled())
+		return;
+
+	WRITE_ONCE(rd->overutilized, status);
+	trace_sched_overutilized_tp(rd, !!status);
+}
+
+static inline void check_update_overutilized_status(struct rq *rq)
+{
+	/*
+	 * overutilized field is used for load balancing decisions only
+	 * if energy aware scheduler is being used
+	 */
+	if (!sched_energy_enabled())
+		return;
+
+	if (!READ_ONCE(rq->rd->overutilized) && cpu_overutilized(rq->cpu))
+		set_rd_overutilized_status(rq->rd, SG_OVERUTILIZED);
 }
 #else
-static inline void update_overutilized_status(struct rq *rq) { }
+static inline void check_update_overutilized_status(struct rq *rq) { }
 #endif
 
 /* Runqueue only has SCHED_IDLE tasks enqueued */
@@ -6779,7 +6799,7 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags)
 	 * and the following generally works well enough in practice.
 	 */
 	if (!task_new)
-		update_overutilized_status(rq);
+		check_update_overutilized_status(rq);
 
 enqueue_throttle:
 	assert_list_leaf_cfs_rq(rq);
@@ -10596,19 +10616,14 @@ static inline void update_sd_lb_stats(struct lb_env *env, struct sd_lb_stats *sd
 		env->fbq_type = fbq_classify_group(&sds->busiest_stat);
 
 	if (!env->sd->parent) {
-		struct root_domain *rd = env->dst_rq->rd;
-
 		/* update overload indicator if we are at root domain */
-		WRITE_ONCE(rd->overload, sg_status & SG_OVERLOAD);
+		WRITE_ONCE(env->dst_rq->rd->overload, sg_status & SG_OVERLOAD);
 
 		/* Update over-utilization (tipping point, U >= 0) indicator */
-		WRITE_ONCE(rd->overutilized, sg_status & SG_OVERUTILIZED);
-		trace_sched_overutilized_tp(rd, sg_status & SG_OVERUTILIZED);
+		set_rd_overutilized_status(env->dst_rq->rd,
+					   sg_status & SG_OVERUTILIZED);
 	} else if (sg_status & SG_OVERUTILIZED) {
-		struct root_domain *rd = env->dst_rq->rd;
-
-		WRITE_ONCE(rd->overutilized, SG_OVERUTILIZED);
-		trace_sched_overutilized_tp(rd, SG_OVERUTILIZED);
+		set_rd_overutilized_status(env->dst_rq->rd, SG_OVERUTILIZED);
 	}
 
 	update_idle_cpu_scan(env, sum_util);
@@ -12609,7 +12624,7 @@ static void task_tick_fair(struct rq *rq, struct task_struct *curr, int queued)
 		task_tick_numa(rq, curr);
 
 	update_misfit_status(curr, rq);
-	update_overutilized_status(task_rq(curr));
+	check_update_overutilized_status(task_rq(curr));
 
 	task_tick_core(rq, curr);
 }

--
2.39.3

maintainer Reviewed-by on v6 1/3
#2 lore raw

Re: [PATCH v6 1/3] sched/fair: Add EAS checks before updating overutilized

Quoted 24 lines · Shrikanth Hegde

Overutilized field of root domain is only used for EAS(energy aware scheduler) to decide whether to do load balance or not. It is not used if EAS not possible.

Currently enqueue_task_fair and task_tick_fair accesses, sometime updates this field. In update_sd_lb_stats it is updated often. This causes cache contention due to true sharing and burns a lot of cycles. overload and overutilized are part of the same cacheline. Updating it often invalidates the cacheline. That causes access to overload to slow down due to false sharing. Hence add EAS check before accessing/updating this field. EAS check is optimized at compile time or it is a static branch. Hence it shouldn't cost much.

With the patch, both enqueue_task_fair and newidle_balance don't show up as hot routines in perf profile.

6.8-rc4:

7.18%  swapper          [kernel.vmlinux]              [k] enqueue_task_fair
6.78%  s                [kernel.vmlinux]              [k] newidle_balance
+patch:
0.14%  swapper          [kernel.vmlinux]              [k] enqueue_task_fair
0.00%  swapper          [kernel.vmlinux]              [k] newidle_balance
Quoted 7 lines

Minor change: trace_sched_overutilized_tp expect that second argument to be bool. So do a int to bool conversion for that.

Fixes 2802bf3cd936 ("sched/fair: Add over-utilization/tipping point indicator")
Quoted 104 lines
---
 kernel/sched/fair.c | 53 +++++++++++++++++++++++++++++----------------
 1 file changed, 34 insertions(+), 19 deletions(-)
kernel/sched/fair.c +34 -20
index 6a16129f9a5c..5aa6e918598c 100644
@@ -6663,22 +6663,42 @@ static inline void hrtick_update(struct rq *rq)
 #ifdef CONFIG_SMP
 static inline bool cpu_overutilized(int cpu)
 {
-       unsigned long rq_util_min = uclamp_rq_get(cpu_rq(cpu), UCLAMP_MIN);
-       unsigned long rq_util_max = uclamp_rq_get(cpu_rq(cpu), UCLAMP_MAX);
+       unsigned long  rq_util_min, rq_util_max;
+
+       if (!sched_energy_enabled())
+               return false;
+
+       rq_util_min = uclamp_rq_get(cpu_rq(cpu), UCLAMP_MIN);
+       rq_util_max = uclamp_rq_get(cpu_rq(cpu), UCLAMP_MAX);
 
        /* Return true only if the utilization doesn't fit CPU's capacity */
        return !util_fits_cpu(cpu_util_cfs(cpu), rq_util_min, rq_util_max, cpu);
 }
 
-static inline void update_overutilized_status(struct rq *rq)
+static inline void set_rd_overutilized_status(struct root_domain *rd,
+                                             unsigned int status)
 {
-       if (!READ_ONCE(rq->rd->overutilized) && cpu_overutilized(rq->cpu)) {
-               WRITE_ONCE(rq->rd->overutilized, SG_OVERUTILIZED);
-               trace_sched_overutilized_tp(rq->rd, SG_OVERUTILIZED);
-       }
+       if (!sched_energy_enabled())
+               return;
+
+       WRITE_ONCE(rd->overutilized, status);
+       trace_sched_overutilized_tp(rd, !!status);
+}
+
+static inline void check_update_overutilized_status(struct rq *rq)
+{
+       /*
+        * overutilized field is used for load balancing decisions only
+        * if energy aware scheduler is being used
+        */
+       if (!sched_energy_enabled())
+               return;
+
+       if (!READ_ONCE(rq->rd->overutilized) && cpu_overutilized(rq->cpu))
+               set_rd_overutilized_status(rq->rd, SG_OVERUTILIZED);
 }
 #else
-static inline void update_overutilized_status(struct rq *rq) { }
+static inline void check_update_overutilized_status(struct rq *rq) { }
 #endif
 
 /* Runqueue only has SCHED_IDLE tasks enqueued */
@@ -6779,7 +6799,7 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags)
         * and the following generally works well enough in practice.
         */
        if (!task_new)
-               update_overutilized_status(rq);
+               check_update_overutilized_status(rq);
 
 enqueue_throttle:
        assert_list_leaf_cfs_rq(rq);
@@ -10596,19 +10616,14 @@ static inline void update_sd_lb_stats(struct lb_env *env, struct sd_lb_stats *sd
                env->fbq_type = fbq_classify_group(&sds->busiest_stat);
 
        if (!env->sd->parent) {
-               struct root_domain *rd = env->dst_rq->rd;
-
                /* update overload indicator if we are at root domain */
-               WRITE_ONCE(rd->overload, sg_status & SG_OVERLOAD);
+               WRITE_ONCE(env->dst_rq->rd->overload, sg_status & SG_OVERLOAD);
 
                /* Update over-utilization (tipping point, U >= 0) indicator */
-               WRITE_ONCE(rd->overutilized, sg_status & SG_OVERUTILIZED);
-               trace_sched_overutilized_tp(rd, sg_status & SG_OVERUTILIZED);
+               set_rd_overutilized_status(env->dst_rq->rd,
+                                          sg_status & SG_OVERUTILIZED);
        } else if (sg_status & SG_OVERUTILIZED) {
-               struct root_domain *rd = env->dst_rq->rd;
-
-               WRITE_ONCE(rd->overutilized, SG_OVERUTILIZED);
-               trace_sched_overutilized_tp(rd, SG_OVERUTILIZED);
+               set_rd_overutilized_status(env->dst_rq->rd, SG_OVERUTILIZED);
        }
 
        update_idle_cpu_scan(env, sum_util);
@@ -12609,7 +12624,7 @@ static void task_tick_fair(struct rq *rq, struct task_struct *curr, int queued)
                task_tick_numa(rq, curr);
 
        update_misfit_status(curr, rq);
-       update_overutilized_status(task_rq(curr));
+       check_update_overutilized_status(task_rq(curr));
 
        task_tick_core(rq, curr);
 }
--

2.39.3

#3 lore raw

[tip: sched/core] sched/fair: Add EAS checks before updating root_domain::overutilized

In reply to Shrikanth Hegde
The following commit has been merged into the sched/core branch of tip:

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

Commit-ID:     be3a51e68f2f1b17250ce40d8872c7645b7a2991
Gitweb:        https://git.kernel.org/tip/be3a51e68f2f1b17250ce40d8872c7645b7a2991
Author:        Shrikanth Hegde <[email protected]>
AuthorDate:    Thu, 07 Mar 2024 14:27:23 +05:30
Committer:     Ingo Molnar <[email protected]>
CommitterDate: Tue, 26 Mar 2024 08:58:59 +01:00

sched/fair: Add EAS checks before updating root_domain::overutilized

root_domain::overutilized is only used for EAS(energy aware scheduler) to decide whether to do load balance or not. It is not used if EAS not possible.

Currently enqueue_task_fair and task_tick_fair accesses, sometime updates this field. In update_sd_lb_stats it is updated often. This causes cache contention due to true sharing and burns a lot of cycles. ::overload and ::overutilized are part of the same cacheline. Updating it often invalidates the cacheline. That causes access to ::overload to slow down due to false sharing. Hence add EAS check before accessing/updating this field. EAS check is optimized at compile time or it is a static branch. Hence it shouldn't cost much.

With the patch, both enqueue_task_fair and newidle_balance don't show up as hot routines in perf profile.

6.8-rc4:

  7.18%  swapper          [kernel.vmlinux]              [k] enqueue_task_fair
  6.78%  s                [kernel.vmlinux]              [k] newidle_balance

+patch:

  0.14%  swapper          [kernel.vmlinux]              [k] enqueue_task_fair
  0.00%  swapper          [kernel.vmlinux]              [k] newidle_balance

While at it: trace_sched_overutilized_tp expect that second argument to be bool. So do a int to bool conversion for that.

---
 kernel/sched/fair.c | 53 ++++++++++++++++++++++++++++----------------
 1 file changed, 34 insertions(+), 19 deletions(-)
kernel/sched/fair.c +34 -19
index dbf4f1c..1afa4f8 100644
@@ -6673,22 +6673,42 @@ static inline void hrtick_update(struct rq *rq)
 #ifdef CONFIG_SMP
 static inline bool cpu_overutilized(int cpu)
 {
-	unsigned long rq_util_min = uclamp_rq_get(cpu_rq(cpu), UCLAMP_MIN);
-	unsigned long rq_util_max = uclamp_rq_get(cpu_rq(cpu), UCLAMP_MAX);
+	unsigned long  rq_util_min, rq_util_max;
+
+	if (!sched_energy_enabled())
+		return false;
+
+	rq_util_min = uclamp_rq_get(cpu_rq(cpu), UCLAMP_MIN);
+	rq_util_max = uclamp_rq_get(cpu_rq(cpu), UCLAMP_MAX);
 
 	/* Return true only if the utilization doesn't fit CPU's capacity */
 	return !util_fits_cpu(cpu_util_cfs(cpu), rq_util_min, rq_util_max, cpu);
 }
 
-static inline void update_overutilized_status(struct rq *rq)
+static inline void set_rd_overutilized_status(struct root_domain *rd,
+					      unsigned int status)
 {
-	if (!READ_ONCE(rq->rd->overutilized) && cpu_overutilized(rq->cpu)) {
-		WRITE_ONCE(rq->rd->overutilized, SG_OVERUTILIZED);
-		trace_sched_overutilized_tp(rq->rd, SG_OVERUTILIZED);
-	}
+	if (!sched_energy_enabled())
+		return;
+
+	WRITE_ONCE(rd->overutilized, status);
+	trace_sched_overutilized_tp(rd, !!status);
+}
+
+static inline void check_update_overutilized_status(struct rq *rq)
+{
+	/*
+	 * overutilized field is used for load balancing decisions only
+	 * if energy aware scheduler is being used
+	 */
+	if (!sched_energy_enabled())
+		return;
+
+	if (!READ_ONCE(rq->rd->overutilized) && cpu_overutilized(rq->cpu))
+		set_rd_overutilized_status(rq->rd, SG_OVERUTILIZED);
 }
 #else
-static inline void update_overutilized_status(struct rq *rq) { }
+static inline void check_update_overutilized_status(struct rq *rq) { }
 #endif
 
 /* Runqueue only has SCHED_IDLE tasks enqueued */
@@ -6789,7 +6809,7 @@ enqueue_task_fair(struct rq *rq, struct task_struct *p, int flags)
 	 * and the following generally works well enough in practice.
 	 */
 	if (!task_new)
-		update_overutilized_status(rq);
+		check_update_overutilized_status(rq);
 
 enqueue_throttle:
 	assert_list_leaf_cfs_rq(rq);
@@ -10630,19 +10650,14 @@ static inline void update_sd_lb_stats(struct lb_env *env, struct sd_lb_stats *sd
 		env->fbq_type = fbq_classify_group(&sds->busiest_stat);
 
 	if (!env->sd->parent) {
-		struct root_domain *rd = env->dst_rq->rd;
-
 		/* update overload indicator if we are at root domain */
-		WRITE_ONCE(rd->overload, sg_status & SG_OVERLOAD);
+		WRITE_ONCE(env->dst_rq->rd->overload, sg_status & SG_OVERLOAD);
 
 		/* Update over-utilization (tipping point, U >= 0) indicator */
-		WRITE_ONCE(rd->overutilized, sg_status & SG_OVERUTILIZED);
-		trace_sched_overutilized_tp(rd, sg_status & SG_OVERUTILIZED);
+		set_rd_overutilized_status(env->dst_rq->rd,
+					   sg_status & SG_OVERUTILIZED);
 	} else if (sg_status & SG_OVERUTILIZED) {
-		struct root_domain *rd = env->dst_rq->rd;
-
-		WRITE_ONCE(rd->overutilized, SG_OVERUTILIZED);
-		trace_sched_overutilized_tp(rd, SG_OVERUTILIZED);
+		set_rd_overutilized_status(env->dst_rq->rd, SG_OVERUTILIZED);
 	}
 
 	update_idle_cpu_scan(env, sum_util);
@@ -12667,7 +12682,7 @@ static void task_tick_fair(struct rq *rq, struct task_struct *curr, int queued)
 		task_tick_numa(rq, curr);
 
 	update_misfit_status(curr, rq);
-	update_overutilized_status(task_rq(curr));
+	check_update_overutilized_status(task_rq(curr));
 
 	task_tick_core(rq, curr);
 }
#4 lore raw

PATCH v6 2/3 sched/fair: Use helper function to access rd->overutilized

In reply to Shrikanth Hegde

Overutilized field is accessed directly in multiple places. So it could use a helper function. That way one might be more informed that it needs to be used only in case of EAS.

No change in functionality intended.

---
 kernel/sched/fair.c | 15 ++++++++++++---
 1 file changed, 12 insertions(+), 3 deletions(-)
kernel/sched/fair.c +12 -3
index 5aa6e918598c..87e08a252f94 100644
@@ -6675,6 +6675,15 @@ static inline bool cpu_overutilized(int cpu)
 	return !util_fits_cpu(cpu_util_cfs(cpu), rq_util_min, rq_util_max, cpu);
 }
 
+/*
+ * Ensure that caller can do EAS. overutilized value
+ * make sense only if EAS is enabled
+ */
+static inline int is_rd_overutilized(struct root_domain *rd)
+{
+	return READ_ONCE(rd->overutilized);
+}
+
 static inline void set_rd_overutilized_status(struct root_domain *rd,
 					      unsigned int status)
 {
@@ -6694,7 +6703,7 @@ static inline void check_update_overutilized_status(struct rq *rq)
 	if (!sched_energy_enabled())
 		return;
 
-	if (!READ_ONCE(rq->rd->overutilized) && cpu_overutilized(rq->cpu))
+	if (!is_rd_overutilized(rq->rd) && cpu_overutilized(rq->cpu))
 		set_rd_overutilized_status(rq->rd, SG_OVERUTILIZED);
 }
 #else
@@ -7980,7 +7989,7 @@ static int find_energy_efficient_cpu(struct task_struct *p, int prev_cpu)
 
 	rcu_read_lock();
 	pd = rcu_dereference(rd->pd);
-	if (!pd || READ_ONCE(rd->overutilized))
+	if (!pd || is_rd_overutilized(rd))
 		goto unlock;
 
 	/*
@@ -10863,7 +10872,7 @@ static struct sched_group *find_busiest_group(struct lb_env *env)
 	if (sched_energy_enabled()) {
 		struct root_domain *rd = env->dst_rq->rd;
 
-		if (rcu_dereference(rd->pd) && !READ_ONCE(rd->overutilized))
+		if (rcu_dereference(rd->pd) && !is_rd_overutilized(rd))
 			goto out_balanced;
 	}
 

--
2.39.3

maintainer Reviewed-by on v6 2/3
#5 lore raw

Re: [PATCH v6 2/3] sched/fair: Use helper function to access rd->overutilized

Quoted 9 lines · Shrikanth Hegde

Overutilized field is accessed directly in multiple places. So it could use a helper function. That way one might be more informed that it needs to be used only in case of EAS.

No change in functionality intended.

Quoted 54 lines
---
 kernel/sched/fair.c | 15 ++++++++++++---
 1 file changed, 12 insertions(+), 3 deletions(-)
kernel/sched/fair.c +12 -4
index 5aa6e918598c..87e08a252f94 100644
@@ -6675,6 +6675,15 @@ static inline bool cpu_overutilized(int cpu)
        return !util_fits_cpu(cpu_util_cfs(cpu), rq_util_min, rq_util_max, cpu);
 }
 
+/*
+ * Ensure that caller can do EAS. overutilized value
+ * make sense only if EAS is enabled
+ */
+static inline int is_rd_overutilized(struct root_domain *rd)
+{
+       return READ_ONCE(rd->overutilized);
+}
+
 static inline void set_rd_overutilized_status(struct root_domain *rd,
                                              unsigned int status)
 {
@@ -6694,7 +6703,7 @@ static inline void check_update_overutilized_status(struct rq *rq)
        if (!sched_energy_enabled())
                return;
 
-       if (!READ_ONCE(rq->rd->overutilized) && cpu_overutilized(rq->cpu))
+       if (!is_rd_overutilized(rq->rd) && cpu_overutilized(rq->cpu))
                set_rd_overutilized_status(rq->rd, SG_OVERUTILIZED);
 }
 #else
@@ -7980,7 +7989,7 @@ static int find_energy_efficient_cpu(struct task_struct *p, int prev_cpu)
 
        rcu_read_lock();
        pd = rcu_dereference(rd->pd);
-       if (!pd || READ_ONCE(rd->overutilized))
+       if (!pd || is_rd_overutilized(rd))
                goto unlock;
 
        /*
@@ -10863,7 +10872,7 @@ static struct sched_group *find_busiest_group(struct lb_env *env)
        if (sched_energy_enabled()) {
                struct root_domain *rd = env->dst_rq->rd;
 
-               if (rcu_dereference(rd->pd) && !READ_ONCE(rd->overutilized))
+               if (rcu_dereference(rd->pd) && !is_rd_overutilized(rd))
                        goto out_balanced;
        }
 
--

2.39.3

#6 lore raw

[tip: sched/core] sched/fair: Introduce is_rd_overutilized() helper function to access root_domain::overutilized

In reply to Shrikanth Hegde
The following commit has been merged into the sched/core branch of tip:

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

Commit-ID:     d0f5d3cefc259f498456338d319098dc84393b24
Gitweb:        https://git.kernel.org/tip/d0f5d3cefc259f498456338d319098dc84393b24
Author:        Shrikanth Hegde <[email protected]>
AuthorDate:    Thu, 07 Mar 2024 14:27:24 +05:30
Committer:     Ingo Molnar <[email protected]>
CommitterDate: Tue, 26 Mar 2024 08:58:59 +01:00

sched/fair: Introduce is_rd_overutilized() helper function to access root_domain::overutilized

The root_domain::overutilized field is READ_ONCE() accessed in multiple places, which could be simplified with a helper function.

This might also make it more apparent that it needs to be used only in case of EAS.

No change in functionality intended.

---
 kernel/sched/fair.c | 15 ++++++++++++---
 1 file changed, 12 insertions(+), 3 deletions(-)
kernel/sched/fair.c +12 -3
index 1afa4f8..24a7530 100644
@@ -6685,6 +6685,15 @@ static inline bool cpu_overutilized(int cpu)
 	return !util_fits_cpu(cpu_util_cfs(cpu), rq_util_min, rq_util_max, cpu);
 }
 
+/*
+ * Ensure that caller can do EAS. overutilized value
+ * make sense only if EAS is enabled
+ */
+static inline int is_rd_overutilized(struct root_domain *rd)
+{
+	return READ_ONCE(rd->overutilized);
+}
+
 static inline void set_rd_overutilized_status(struct root_domain *rd,
 					      unsigned int status)
 {
@@ -6704,7 +6713,7 @@ static inline void check_update_overutilized_status(struct rq *rq)
 	if (!sched_energy_enabled())
 		return;
 
-	if (!READ_ONCE(rq->rd->overutilized) && cpu_overutilized(rq->cpu))
+	if (!is_rd_overutilized(rq->rd) && cpu_overutilized(rq->cpu))
 		set_rd_overutilized_status(rq->rd, SG_OVERUTILIZED);
 }
 #else
@@ -7990,7 +7999,7 @@ static int find_energy_efficient_cpu(struct task_struct *p, int prev_cpu)
 
 	rcu_read_lock();
 	pd = rcu_dereference(rd->pd);
-	if (!pd || READ_ONCE(rd->overutilized))
+	if (!pd || is_rd_overutilized(rd))
 		goto unlock;
 
 	/*
@@ -10897,7 +10906,7 @@ static struct sched_group *sched_balance_find_src_group(struct lb_env *env)
 	if (sched_energy_enabled()) {
 		struct root_domain *rd = env->dst_rq->rd;
 
-		if (rcu_dereference(rd->pd) && !READ_ONCE(rd->overutilized))
+		if (rcu_dereference(rd->pd) && !is_rd_overutilized(rd))
 			goto out_balanced;
 	}
 
#7 lore raw

PATCH v6 3/3 sched/fair: Combine EAS check with overutilized access

In reply to Shrikanth Hegde

Access to overutilized is always used with sched_energy_enabled in the pattern:

if (sched_energy_enabled && !overutilized)
do something

So modify the helper function to return this pattern. This is more readable code as it would say, do something when root domain is not overutilized.

No change in functionality intended.

---
 kernel/sched/fair.c | 24 +++++++++---------------
 1 file changed, 9 insertions(+), 15 deletions(-)
kernel/sched/fair.c +9 -15
index 87e08a252f94..bcda18a2ccfe 100644
@@ -6676,12 +6676,11 @@ static inline bool cpu_overutilized(int cpu)
 }
 
 /*
- * Ensure that caller can do EAS. overutilized value
- * make sense only if EAS is enabled
+ * overutilized value make sense only if EAS is enabled
  */
-static inline int is_rd_overutilized(struct root_domain *rd)
+static inline int is_rd_not_overutilized(struct root_domain *rd)
 {
-	return READ_ONCE(rd->overutilized);
+	return sched_energy_enabled() && !READ_ONCE(rd->overutilized);
 }
 
 static inline void set_rd_overutilized_status(struct root_domain *rd,
@@ -6700,10 +6699,8 @@ static inline void check_update_overutilized_status(struct rq *rq)
 	 * overutilized field is used for load balancing decisions only
 	 * if energy aware scheduler is being used
 	 */
-	if (!sched_energy_enabled())
-		return;
 
-	if (!is_rd_overutilized(rq->rd) && cpu_overutilized(rq->cpu))
+	if (is_rd_not_overutilized(rq->rd) && cpu_overutilized(rq->cpu))
 		set_rd_overutilized_status(rq->rd, SG_OVERUTILIZED);
 }
 #else
@@ -7989,7 +7986,7 @@ static int find_energy_efficient_cpu(struct task_struct *p, int prev_cpu)
 
 	rcu_read_lock();
 	pd = rcu_dereference(rd->pd);
-	if (!pd || is_rd_overutilized(rd))
+	if (!pd)
 		goto unlock;
 
 	/*
@@ -8192,7 +8189,7 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int wake_flags)
 		    cpumask_test_cpu(cpu, p->cpus_ptr))
 			return cpu;
 
-		if (sched_energy_enabled()) {
+		if (is_rd_not_overutilized(this_rq()->rd)) {
 			new_cpu = find_energy_efficient_cpu(p, prev_cpu);
 			if (new_cpu >= 0)
 				return new_cpu;
@@ -10869,12 +10866,9 @@ static struct sched_group *find_busiest_group(struct lb_env *env)
 	if (busiest->group_type == group_misfit_task)
 		goto force_balance;
 
-	if (sched_energy_enabled()) {
-		struct root_domain *rd = env->dst_rq->rd;
-
-		if (rcu_dereference(rd->pd) && !is_rd_overutilized(rd))
-			goto out_balanced;
-	}
+	if (is_rd_not_overutilized(env->dst_rq->rd) &&
+	    rcu_dereference(env->dst_rq->rd->pd))
+		goto out_balanced;
 
 	/* ASYM feature bypasses nice load balance check */
 	if (busiest->group_type == group_asym_packing)

--
2.39.3

maintainer Reviewed-by on v6 3/3
#8 lore raw

Re: [PATCH v6 3/3] sched/fair: Combine EAS check with overutilized access

Quoted 7 lines · Shrikanth Hegde

Access to overutilized is always used with sched_energy_enabled in the pattern:

if (sched_energy_enabled && !overutilized)
do something

Quoted 8 lines

So modify the helper function to return this pattern. This is more readable code as it would say, do something when root domain is not overutilized.

No change in functionality intended.

Quoted 73 lines
---
 kernel/sched/fair.c | 24 +++++++++---------------
 1 file changed, 9 insertions(+), 15 deletions(-)
kernel/sched/fair.c +9 -16
index 87e08a252f94..bcda18a2ccfe 100644
@@ -6676,12 +6676,11 @@ static inline bool cpu_overutilized(int cpu)
 }
 
 /*
- * Ensure that caller can do EAS. overutilized value
- * make sense only if EAS is enabled
+ * overutilized value make sense only if EAS is enabled
  */
-static inline int is_rd_overutilized(struct root_domain *rd)
+static inline int is_rd_not_overutilized(struct root_domain *rd)
 {
-       return READ_ONCE(rd->overutilized);
+       return sched_energy_enabled() && !READ_ONCE(rd->overutilized);
 }
 
 static inline void set_rd_overutilized_status(struct root_domain *rd,
@@ -6700,10 +6699,8 @@ static inline void check_update_overutilized_status(struct rq *rq)
         * overutilized field is used for load balancing decisions only
         * if energy aware scheduler is being used
         */
-       if (!sched_energy_enabled())
-               return;
 
-       if (!is_rd_overutilized(rq->rd) && cpu_overutilized(rq->cpu))
+       if (is_rd_not_overutilized(rq->rd) && cpu_overutilized(rq->cpu))
                set_rd_overutilized_status(rq->rd, SG_OVERUTILIZED);
 }
 #else
@@ -7989,7 +7986,7 @@ static int find_energy_efficient_cpu(struct task_struct *p, int prev_cpu)
 
        rcu_read_lock();
        pd = rcu_dereference(rd->pd);
-       if (!pd || is_rd_overutilized(rd))
+       if (!pd)
                goto unlock;
 
        /*
@@ -8192,7 +8189,7 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int wake_flags)
                    cpumask_test_cpu(cpu, p->cpus_ptr))
                        return cpu;
 
-               if (sched_energy_enabled()) {
+               if (is_rd_not_overutilized(this_rq()->rd)) {
                        new_cpu = find_energy_efficient_cpu(p, prev_cpu);
                        if (new_cpu >= 0)
                                return new_cpu;
@@ -10869,12 +10866,9 @@ static struct sched_group *find_busiest_group(struct lb_env *env)
        if (busiest->group_type == group_misfit_task)
                goto force_balance;
 
-       if (sched_energy_enabled()) {
-               struct root_domain *rd = env->dst_rq->rd;
-
-               if (rcu_dereference(rd->pd) && !is_rd_overutilized(rd))
-                       goto out_balanced;
-       }
+       if (is_rd_not_overutilized(env->dst_rq->rd) &&
+           rcu_dereference(env->dst_rq->rd->pd))
+               goto out_balanced;
 
        /* ASYM feature bypasses nice load balance check */
        if (busiest->group_type == group_asym_packing)
--

2.39.3

#9 lore raw

Re: [PATCH v6 3/3] sched/fair: Combine EAS check with overutilized access

In reply to Shrikanth Hegde

kernel/sched/fair.c static inline bool cpu_overutilized(int cpu) · line 6684

Quoted 11 lines · Shrikanth Hegde

/*
- * Ensure that caller can do EAS. overutilized value
- * make sense only if EAS is enabled
+ * overutilized value make sense only if EAS is enabled */

kernel/sched/fair.c +2 -2
-static inline int is_rd_overutilized(struct root_domain *rd)
+static inline int is_rd_not_overutilized(struct root_domain *rd)
 {
-	return READ_ONCE(rd->overutilized);
+	return sched_energy_enabled() && !READ_ONCE(rd->overutilized);
 }

While adding the sched_energy_enabled() condition looks OK, the _not prefix This is silly: putting logical operators into functions names is far less readable than a !fn()...

kernel/sched/fair.c static inline void check_update_overutilized_status(struct rq *rq) · line 6703

Quoted 2 lines
kernel/sched/fair.c +1 -1
-	if (!is_rd_overutilized(rq->rd) && cpu_overutilized(rq->cpu))
+	if (is_rd_not_overutilized(rq->rd) && cpu_overutilized(rq->cpu))

Especially since we already have cpu_overutilized(). It's far more coherent to have the same basic attribute functions and put any negation into *actual* logical operators.

Thanks,

Ingo

maintainer
#10 lore raw

Re: [PATCH v6 3/3] sched/fair: Combine EAS check with overutilized access

Quoted 20 lines · Ingo Molnar

kernel/sched/fair.c static inline bool cpu_overutilized(int cpu) · line 6684

Quoted 11 lines · Shrikanth Hegde

/*
- * Ensure that caller can do EAS. overutilized value
- * make sense only if EAS is enabled
+ * overutilized value make sense only if EAS is enabled */

kernel/sched/fair.c +2 -2
-static inline int is_rd_overutilized(struct root_domain *rd)
+static inline int is_rd_not_overutilized(struct root_domain *rd)
 {
-     return READ_ONCE(rd->overutilized);
+     return sched_energy_enabled() && !READ_ONCE(rd->overutilized);
 }

While adding the sched_energy_enabled() condition looks OK, the _not prefix This is silly: putting logical operators into functions names is far less readable than a !fn()...

Quoted 6 lines

kernel/sched/fair.c static inline void check_update_overutilized_status(struct rq *rq) · line 6703

Quoted 2 lines
kernel/sched/fair.c +1 -1
-     if (!is_rd_overutilized(rq->rd) && cpu_overutilized(rq->cpu))
+     if (is_rd_not_overutilized(rq->rd) && cpu_overutilized(rq->cpu))

Especially since we already have cpu_overutilized(). It's far more coherent to have the same basic attribute functions and put any negation into *actual* logical operators.

I was concerned by the || in this case that could defeat the purpose of sched_energy_enabled() but it will return early anyway

return !sched_energy_enabled() || READ_ONCE(rd->overutilized);

Quoted 4 lines

Thanks,

Ingo

#11 lore raw

Re: [PATCH v6 3/3] sched/fair: Combine EAS check with overutilized access

Quoted 28 lines · Vincent Guittot
Quoted 20 lines · Ingo Molnar

Review

Quoted 11 lines · Shrikanth Hegde

/*
- * Ensure that caller can do EAS. overutilized value
- * make sense only if EAS is enabled
+ * overutilized value make sense only if EAS is enabled */

(quoted diff) +2 -2
-static inline int is_rd_overutilized(struct root_domain *rd)
+static inline int is_rd_not_overutilized(struct root_domain *rd)
 {
-     return READ_ONCE(rd->overutilized);
+     return sched_energy_enabled() && !READ_ONCE(rd->overutilized);
 }

While adding the sched_energy_enabled() condition looks OK, the _not prefix This is silly: putting logical operators into functions names is far less readable than a !fn()...

Quoted 6 lines

Review

Quoted 2 lines
(quoted diff) +1 -1
-     if (!is_rd_overutilized(rq->rd) && cpu_overutilized(rq->cpu))
+     if (is_rd_not_overutilized(rq->rd) && cpu_overutilized(rq->cpu))

Especially since we already have cpu_overutilized(). It's far more coherent to have the same basic attribute functions and put any negation into *actual* logical operators.

Quoted 5 lines

I was concerned by the || in this case that could defeat the purpose of sched_energy_enabled() but it will return early anyway

return !sched_energy_enabled() || READ_ONCE(rd->overutilized);

I think this would work.

Quoted 5 lines
Quoted 5 lines

Thanks,

Ingo

If EAS - false, then is_rd_overutilized -> would be true always and all users of it do !is_rd_overutilized(). so No operation. If EAS - true, it reads rd->overutilized value.

Does this look correct?
-------------------------------------------------------------------------------------
From 3adc0d58f87d8a2e96196a0f47bcd0d2afd057ae Mon Sep 17 00:00:00 2001
From: Shrikanth Hegde <[email protected]>
Date: Wed, 6 Mar 2024 03:58:58 -0500
Subject: [PATCH v7 3/3] sched/fair: Combine EAS check with overutilized access

Access to overutilized is always used with sched_energy_enabled in the pattern:

if (sched_energy_enabled && !overutilized)
do something

So modify the helper function to return this pattern. This is more readable code as it would say, do something when root domain is not overutilized. This function always return true when EAS is disabled.

No change in functionality intended.

---
 kernel/sched/fair.c | 20 +++++++-------------
 1 file changed, 7 insertions(+), 13 deletions(-)
kernel/sched/fair.c +7 -13
index 24a7530a7d3f..e222e3ad4cfe 100644
@@ -6686,12 +6686,11 @@ static inline bool cpu_overutilized(int cpu)
 }
 
 /*
- * Ensure that caller can do EAS. overutilized value
- * make sense only if EAS is enabled
+ * overutilized value make sense only if EAS is enabled
  */
 static inline int is_rd_overutilized(struct root_domain *rd)
 {
-	return READ_ONCE(rd->overutilized);
+	return !sched_energy_enabled() || READ_ONCE(rd->overutilized);
 }
 
 static inline void set_rd_overutilized_status(struct root_domain *rd,
@@ -6710,8 +6709,6 @@ static inline void check_update_overutilized_status(struct rq *rq)
 	 * overutilized field is used for load balancing decisions only
 	 * if energy aware scheduler is being used
 	 */
-	if (!sched_energy_enabled())
-		return;
 
 	if (!is_rd_overutilized(rq->rd) && cpu_overutilized(rq->cpu))
 		set_rd_overutilized_status(rq->rd, SG_OVERUTILIZED);
@@ -7999,7 +7996,7 @@ static int find_energy_efficient_cpu(struct task_struct *p, int prev_cpu)
 
 	rcu_read_lock();
 	pd = rcu_dereference(rd->pd);
-	if (!pd || is_rd_overutilized(rd))
+	if (!pd)
 		goto unlock;
 
 	/*
@@ -8202,7 +8199,7 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int wake_flags)
 		    cpumask_test_cpu(cpu, p->cpus_ptr))
 			return cpu;
 
-		if (sched_energy_enabled()) {
+		if (!is_rd_overutilized(this_rq()->rd)) {
 			new_cpu = find_energy_efficient_cpu(p, prev_cpu);
 			if (new_cpu >= 0)
 				return new_cpu;
@@ -10903,12 +10900,9 @@ static struct sched_group *sched_balance_find_src_group(struct lb_env *env)
 	if (busiest->group_type == group_misfit_task)
 		goto force_balance;
 
-	if (sched_energy_enabled()) {
-		struct root_domain *rd = env->dst_rq->rd;
-
-		if (rcu_dereference(rd->pd) && !is_rd_overutilized(rd))
-			goto out_balanced;
-	}
+	if (!is_rd_overutilized(env->dst_rq->rd) &&
+	    rcu_dereference(env->dst_rq->rd->pd))
+		goto out_balanced;
 
 	/* ASYM feature bypasses nice load balance check */
 	if (busiest->group_type == group_asym_packing)

--
2.39.3

maintainer
#12 lore raw

Re: [PATCH v6 3/3] sched/fair: Combine EAS check with overutilized access

Quoted 44 lines · Shrikanth Hegde
Quoted 28 lines · Vincent Guittot
Quoted 20 lines · Ingo Molnar

kernel/sched/fair.c static inline bool cpu_overutilized(int cpu) · line 6694

Quoted 11 lines · Shrikanth Hegde

/*
- * Ensure that caller can do EAS. overutilized value
- * make sense only if EAS is enabled
+ * overutilized value make sense only if EAS is enabled */

kernel/sched/fair.c +2 -2
-static inline int is_rd_overutilized(struct root_domain *rd)
+static inline int is_rd_not_overutilized(struct root_domain *rd)
 {
-     return READ_ONCE(rd->overutilized);
+     return sched_energy_enabled() && !READ_ONCE(rd->overutilized);
 }

While adding the sched_energy_enabled() condition looks OK, the _not prefix This is silly: putting logical operators into functions names is far less readable than a !fn()...

Quoted 6 lines

Review

Quoted 2 lines
(quoted diff) +1 -1
-     if (!is_rd_overutilized(rq->rd) && cpu_overutilized(rq->cpu))
+     if (is_rd_not_overutilized(rq->rd) && cpu_overutilized(rq->cpu))

Especially since we already have cpu_overutilized(). It's far more coherent to have the same basic attribute functions and put any negation into *actual* logical operators.

Quoted 5 lines

I was concerned by the || in this case that could defeat the purpose of sched_energy_enabled() but it will return early anyway

return !sched_energy_enabled() || READ_ONCE(rd->overutilized);

I think this would work.

Quoted 4 lines
Quoted 4 lines

Thanks,

Quoted 7 lines
Quoted 1 line
Quoted 1 line

Ingo

If EAS - false, then is_rd_overutilized -> would be true always and all users of it do !is_rd_overutilized(). so No operation. If EAS - true, it reads rd->overutilized value.

Does this look correct?

yes looks good to me

Quoted 88 lines

-------------------------------------------------------------------------------------
From 3adc0d58f87d8a2e96196a0f47bcd0d2afd057ae Mon Sep 17 00:00:00 2001
From: Shrikanth Hegde <[email protected]>
Date: Wed, 6 Mar 2024 03:58:58 -0500
Subject: [PATCH v7 3/3] sched/fair: Combine EAS check with overutilized access

Access to overutilized is always used with sched_energy_enabled in the pattern:

if (sched_energy_enabled && !overutilized)
do something

So modify the helper function to return this pattern. This is more readable code as it would say, do something when root domain is not overutilized. This function always return true when EAS is disabled.

No change in functionality intended.

---
 kernel/sched/fair.c | 20 +++++++-------------
 1 file changed, 7 insertions(+), 13 deletions(-)
kernel/sched/fair.c +7 -14
index 24a7530a7d3f..e222e3ad4cfe 100644
@@ -6686,12 +6686,11 @@ static inline bool cpu_overutilized(int cpu)
 }
 
 /*
- * Ensure that caller can do EAS. overutilized value
- * make sense only if EAS is enabled
+ * overutilized value make sense only if EAS is enabled
  */
 static inline int is_rd_overutilized(struct root_domain *rd)
 {
-       return READ_ONCE(rd->overutilized);
+       return !sched_energy_enabled() || READ_ONCE(rd->overutilized);
 }
 
 static inline void set_rd_overutilized_status(struct root_domain *rd,
@@ -6710,8 +6709,6 @@ static inline void check_update_overutilized_status(struct rq *rq)
         * overutilized field is used for load balancing decisions only
         * if energy aware scheduler is being used
         */
-       if (!sched_energy_enabled())
-               return;
 
        if (!is_rd_overutilized(rq->rd) && cpu_overutilized(rq->cpu))
                set_rd_overutilized_status(rq->rd, SG_OVERUTILIZED);
@@ -7999,7 +7996,7 @@ static int find_energy_efficient_cpu(struct task_struct *p, int prev_cpu)
 
        rcu_read_lock();
        pd = rcu_dereference(rd->pd);
-       if (!pd || is_rd_overutilized(rd))
+       if (!pd)
                goto unlock;
 
        /*
@@ -8202,7 +8199,7 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int wake_flags)
                    cpumask_test_cpu(cpu, p->cpus_ptr))
                        return cpu;
 
-               if (sched_energy_enabled()) {
+               if (!is_rd_overutilized(this_rq()->rd)) {
                        new_cpu = find_energy_efficient_cpu(p, prev_cpu);
                        if (new_cpu >= 0)
                                return new_cpu;
@@ -10903,12 +10900,9 @@ static struct sched_group *sched_balance_find_src_group(struct lb_env *env)
        if (busiest->group_type == group_misfit_task)
                goto force_balance;
 
-       if (sched_energy_enabled()) {
-               struct root_domain *rd = env->dst_rq->rd;
-
-               if (rcu_dereference(rd->pd) && !is_rd_overutilized(rd))
-                       goto out_balanced;
-       }
+       if (!is_rd_overutilized(env->dst_rq->rd) &&
+           rcu_dereference(env->dst_rq->rd->pd))
+               goto out_balanced;
 
        /* ASYM feature bypasses nice load balance check */
        if (busiest->group_type == group_asym_packing)
--

2.39.3

#13 lore raw

Re: [PATCH v6 3/3] sched/fair: Combine EAS check with overutilized access

Quoted 31 lines · Vincent Guittot
Quoted 30 lines · Shrikanth Hegde
Quoted 15 lines
Quoted 8 lines

While adding the sched_energy_enabled() condition looks OK, the _not prefix This is silly: putting logical operators into functions names is far less readable than a !fn()...

Quoted 6 lines

Review

Quoted 2 lines
(quoted diff) +1 -1
-     if (!is_rd_overutilized(rq->rd) && cpu_overutilized(rq->cpu))
+     if (is_rd_not_overutilized(rq->rd) && cpu_overutilized(rq->cpu))

Especially since we already have cpu_overutilized(). It's far more coherent to have the same basic attribute functions and put any negation into *actual* logical operators.

Quoted 5 lines

I was concerned by the || in this case that could defeat the purpose of sched_energy_enabled() but it will return early anyway

return !sched_energy_enabled() || READ_ONCE(rd->overutilized);

I think this would work.

Quoted 5 lines
Quoted 5 lines

Thanks,

Ingo

Quoted 7 lines
Quoted 4 lines

If EAS - false, then is_rd_overutilized -> would be true always and all users of it do !is_rd_overutilized(). so No operation. If EAS - true, it reads rd->overutilized value.

Does this look correct?

yes looks good to me

Thank you. Let me send it as a patch

Quoted 88 lines
Quoted 88 lines

-------------------------------------------------------------------------------------
From 3adc0d58f87d8a2e96196a0f47bcd0d2afd057ae Mon Sep 17 00:00:00 2001
From: Shrikanth Hegde <[email protected]>
Date: Wed, 6 Mar 2024 03:58:58 -0500
Subject: [PATCH v7 3/3] sched/fair: Combine EAS check with overutilized access

Access to overutilized is always used with sched_energy_enabled in the pattern:

if (sched_energy_enabled && !overutilized)
do something

So modify the helper function to return this pattern. This is more readable code as it would say, do something when root domain is not overutilized. This function always return true when EAS is disabled.

No change in functionality intended.

---
 kernel/sched/fair.c | 20 +++++++-------------
 1 file changed, 7 insertions(+), 13 deletions(-)
kernel/sched/fair.c +7 -14
index 24a7530a7d3f..e222e3ad4cfe 100644
@@ -6686,12 +6686,11 @@ static inline bool cpu_overutilized(int cpu)
 }
 
 /*
- * Ensure that caller can do EAS. overutilized value
- * make sense only if EAS is enabled
+ * overutilized value make sense only if EAS is enabled
  */
 static inline int is_rd_overutilized(struct root_domain *rd)
 {
-       return READ_ONCE(rd->overutilized);
+       return !sched_energy_enabled() || READ_ONCE(rd->overutilized);
 }
 
 static inline void set_rd_overutilized_status(struct root_domain *rd,
@@ -6710,8 +6709,6 @@ static inline void check_update_overutilized_status(struct rq *rq)
         * overutilized field is used for load balancing decisions only
         * if energy aware scheduler is being used
         */
-       if (!sched_energy_enabled())
-               return;
 
        if (!is_rd_overutilized(rq->rd) && cpu_overutilized(rq->cpu))
                set_rd_overutilized_status(rq->rd, SG_OVERUTILIZED);
@@ -7999,7 +7996,7 @@ static int find_energy_efficient_cpu(struct task_struct *p, int prev_cpu)
 
        rcu_read_lock();
        pd = rcu_dereference(rd->pd);
-       if (!pd || is_rd_overutilized(rd))
+       if (!pd)
                goto unlock;
 
        /*
@@ -8202,7 +8199,7 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int wake_flags)
                    cpumask_test_cpu(cpu, p->cpus_ptr))
                        return cpu;
 
-               if (sched_energy_enabled()) {
+               if (!is_rd_overutilized(this_rq()->rd)) {
                        new_cpu = find_energy_efficient_cpu(p, prev_cpu);
                        if (new_cpu >= 0)
                                return new_cpu;
@@ -10903,12 +10900,9 @@ static struct sched_group *sched_balance_find_src_group(struct lb_env *env)
        if (busiest->group_type == group_misfit_task)
                goto force_balance;
 
-       if (sched_energy_enabled()) {
-               struct root_domain *rd = env->dst_rq->rd;
-
-               if (rcu_dereference(rd->pd) && !is_rd_overutilized(rd))
-                       goto out_balanced;
-       }
+       if (!is_rd_overutilized(env->dst_rq->rd) &&
+           rcu_dereference(env->dst_rq->rd->pd))
+               goto out_balanced;
 
        /* ASYM feature bypasses nice load balance check */
        if (busiest->group_type == group_asym_packing)
--

2.39.3

reviewer Reviewed-by on the v6 series
#14 lore raw
In reply to Shrikanth Hegde
Quoted 13 lines · Shrikanth Hegde

When running an ISV workload on a large system (240 Cores, SMT8), it was observed from perf profile that newidle_balance and enqueue_task_fair were consuming more cycles. Perf annotate showed that most of the time was spent on accessing overutilized field of root domain. Similar perf profile was simulated by making some changes to stress-ng --wait. Both newidle_balance and enqueue_task_fair consume close to 5-7%.

Overutilized was added for EAS(Energy aware scheduler) to decide whether to do load balance or not. Simultaneous access to overutilized by multiple CPUs lead cache invalidations due to true sharing. Updating overutilized is not required for non-EAS platforms. Since overutilized and overload are part of the same cacheline, there is false sharing as well.

Quoted 10 lines

Patch 1/3 - Main patch. It helps in reducing the above said issue. Both the functions don't show up in the profile. With patch comparison is in changelog. With the patch stated problem in the ISV workload also got solved and throughput has improved.
Patch 2/3 - Code refactoring to use the helper function instead of direct access of the field. Keeping this patch so patch 3/3 becomes easier to understand. Depends on 1/3.
Patch 3/3 - Refactoring the code since most of the patterns are observed are eas && !overutilzed. Changed the helper function accordingly. Depends on 2/3.