Skip to contentSkip to conversation

sched_ext: split global idle cpumask into per-NUMA cpumasks

PATCH Superseded Andrea Righi 1 participant 8 messages Dec 9, 2024

Subsystems (2)

v6 of 12 Superseded Latest is v12 →

lore
Share & tools
lore

Andrea Righi

author reviewer
#0 lore raw

PATCH v6 1/1 sched_ext: split global idle cpumask into per-NUMA cpumasks

= Overview =

As discussed during the sched_ext office hours, using a global cpumask to keep track of the idle CPUs can be inefficient and it may not scale really well on large NUMA systems.

Therefore, split the idle cpumask into multiple per-NUMA node cpumasks to improve scalability and performance on such large systems.

Scalability issues seem to be more noticeable on Intel Sapphire Rapids dual-socket architectures.

= Test =

Hardware:
- System: DGX B200
- CPUs: 224 SMT threads (112 physical cores)
- Processor: INTEL(R) XEON(R) PLATINUM 8570
- 2 NUMA nodes

Scheduler:
- scx_simple [1] (so that we can focus at the built-in idle selection policy and not at the scheduling policy itself)

Test:
- Run a parallel kernel build make -j $(nproc) and measure the average elapsed time over 10 runs:

avg time | stdev
---------+------

 before:   52.431s | 2.895
  after:   50.342s | 2.895

= Conclusion =

Splitting the global cpumask into multiple per-NUMA cpumasks helped to achieve a speedup of approximately +4% with this particular architecture and test case.

I've repeated the same test on a DGX-1 (40 physical cores, Intel Xeon E5-2698 v4 @ 2.20GHz, 2 NUMA nodes) and I didn't observe any measurable difference.

In general, on smaller systems, I haven't noticed any measurable regressions or improvements with the same test (parallel kernel build) and scheduler (scx_simple).

NOTE: splitting the global cpumask into multiple cpumasks may increase the overhead of scx_bpf_pick_idle_cpu() or ops.select_cpu() (for schedulers relying on the built-in CPU idle selection policy) in presence of multiple NUMA nodes, particularly under high system load, since we may have to access multiple cpumasks to find an idle CPU.

However, this increased overhead seems to be highly compensated by a lower overhead when updating the idle state (__scx_update_idle()) and by the fact that CPUs are more likely operating within their local idle cpumask, reducing the stress on the cache coherency protocol.

= References =

[1] https://github.com/sched-ext/scx/blob/main/scheds/c/scx_simple.bpf.c

ChangeLog v5 -> v6:
- refactor patch set to introduce SCX_OPS_NODE_BUILTIN_IDLE before the per-node cpumasks
- move idle CPU selection policy to a separate file (ext_idle.c) (no functional change, just some code shuffling)

ChangeLog v4 -> v5:
- introduce new scx_bpf_cpu_to_node() kfunc
- provide __COMPAT_*() helpers for the new kfunc's

ChangeLog v3 -> v4:
- introduce SCX_OPS_NODE_BUILTIN_IDLE to select multiple per-node cpumasks or single flat cpumask
- introduce new kfuncs to access per-node idle cpumasks information
- use for_each_numa_hop_mask() to traverse NUMA nodes in increasing distance
- dropped nodemask helpers (not needed anymore)
- rebase to sched_ext/for-6.14

ChangeLog v2 -> v3:
- introduce for_each_online_node_wrap()
- re-introduce cpumask_intersects() in test_and_clear_cpu_idle() (to reduce memory writes / cache coherence pressure)
- get rid of the redundant scx_selcpu_topo_numa logic
[test results are pretty much identical, so I haven't updated them from v2]

ChangeLog v1 -> v2:
- renamed for_each_node_mask|state_from() -> for_each_node_mask|state_wrap()
- misc cpumask optimizations (thanks to Yury)

Andrea Righi (6):
sched_ext: Introduce SCX_OPS_NODE_BUILTIN_IDLE
sched_ext: Introduce per-node idle cpumasks
sched_ext: Get rid of the scx_selcpu_topo_numa logic
sched_ext: Use SCX_OPS_NODE_BUILTIN_IDLE
sched_ext: Introduce NUMA aware idle cpu kfunc helpers
sched_ext: Move built-in idle CPU selection policy to a separate file

 MAINTAINERS                              |   1 +
 kernel/sched/ext.c                       | 744 ++-------------------------
 kernel/sched/ext_idle.c                  | 845 +++++++++++++++++++++++++++++++
 tools/sched_ext/include/scx/common.bpf.h |   4 +
 tools/sched_ext/include/scx/compat.bpf.h |  19 +
 5 files changed, 912 insertions(+), 701 deletions(-)
 create mode 100644 kernel/sched/ext_idle.c
author reviewer
#1 lore raw

PATCH 1/6 sched_ext: Introduce SCX_OPS_NODE_BUILTIN_IDLE

Add a flag (SCX_OPS_NODE_BUILTIN_IDLE) to toggle between a global flat idle cpumask and multiple per-node CPU masks.

---
 kernel/sched/ext.c | 7 +++++++
 1 file changed, 7 insertions(+)
kernel/sched/ext.c +7 -0
index 71342f3719c1..3d13e7975abd 100644
@@ -122,6 +122,12 @@ enum scx_ops_flags {
 	 */
 	SCX_OPS_SWITCH_PARTIAL	= 1LLU << 3,
 
+	/*
+	 * If set, enable per-node idle cpumasks. If clear, use a single global
+	 * flat idle cpumask.
+	 */
+	SCX_OPS_BUILTIN_IDLE_PER_NODE = 1LLU << 4,
+
 	/*
 	 * CPU cgroup support flags
 	 */
@@ -131,6 +137,7 @@ enum scx_ops_flags {
 				  SCX_OPS_ENQ_LAST |
 				  SCX_OPS_ENQ_EXITING |
 				  SCX_OPS_SWITCH_PARTIAL |
+				  SCX_OPS_BUILTIN_IDLE_PER_NODE |
 				  SCX_OPS_HAS_CGROUP_WEIGHT,
 };
 
Signature · 1 line
2.47.1
author reviewer
#2 lore raw

PATCH 2/6 sched_ext: Introduce per-node idle cpumasks

In reply to Andrea Righi

Using a single global idle mask can lead to inefficiencies and a lot of stress on the cache coherency protocol on large systems with multiple NUMA nodes, since all the CPUs can create a really intense read/write activity on the single global cpumask.

Therefore, split the global cpumask into multiple per-NUMA node cpumasks to improve scalability and performance on large systems.

The concept is that each cpumask will track only the idle CPUs within its corresponding NUMA node, treating CPUs in other NUMA nodes as busy. In this way concurrent access to the idle cpumask will be restricted within each NUMA node.

NOTE: with per-node idle cpumasks enabled scx_bpf_get_idle_cpu/smtmask() are returning the cpumask of the current NUMA node, instead of a single cpumask for all the CPUs.

---
 kernel/sched/ext.c | 221 ++++++++++++++++++++++++++++++++++-----------
 1 file changed, 166 insertions(+), 55 deletions(-)
kernel/sched/ext.c +166 -55
index 3d13e7975abd..0b9b6627515b 100644
@@ -940,8 +940,60 @@ static struct delayed_work scx_watchdog_work;
 static struct {
 	cpumask_var_t cpu;
 	cpumask_var_t smt;
-} idle_masks CL_ALIGNED_IF_ONSTACK;
+} **idle_masks CL_ALIGNED_IF_ONSTACK;
 
+static struct cpumask *get_idle_cpumask_node(int node)
+{
+	return idle_masks[node]->cpu;
+}
+
+static struct cpumask *get_idle_smtmask_node(int node)
+{
+	return idle_masks[node]->smt;
+}
+
+static struct cpumask *get_curr_idle_cpumask(void)
+{
+	int node = cpu_to_node(smp_processor_id());
+
+	return get_idle_cpumask_node(node);
+}
+
+static struct cpumask *get_curr_idle_smtmask(void)
+{
+	int node = cpu_to_node(smp_processor_id());
+
+	if (sched_smt_active())
+		return get_idle_smtmask_node(node);
+	else
+		return get_idle_cpumask_node(node);
+}
+
+static void idle_masks_init(void)
+{
+	int node;
+
+	idle_masks = kcalloc(num_possible_nodes(), sizeof(*idle_masks), GFP_KERNEL);
+	BUG_ON(!idle_masks);
+
+	for_each_node_state(node, N_POSSIBLE) {
+		idle_masks[node] = kzalloc_node(sizeof(**idle_masks), GFP_KERNEL, node);
+		BUG_ON(!idle_masks[node]);
+
+		BUG_ON(!alloc_cpumask_var_node(&idle_masks[node]->cpu, GFP_KERNEL, node));
+		BUG_ON(!alloc_cpumask_var_node(&idle_masks[node]->smt, GFP_KERNEL, node));
+	}
+}
+#else	/* !CONFIG_SMP */
+static struct cpumask *get_curr_idle_cpumask(void)
+{
+	return cpu_none_mask;
+}
+
+static struct cpumask *get_curr_idle_smtmask(void)
+{
+	return cpu_none_mask;
+}
 #endif	/* CONFIG_SMP */
 
 /* for %SCX_KICK_WAIT */
@@ -3173,6 +3225,9 @@ bool scx_prio_less(const struct task_struct *a, const struct task_struct *b,
 
 static bool test_and_clear_cpu_idle(int cpu)
 {
+	int node = cpu_to_node(cpu);
+	struct cpumask *idle_cpu = get_idle_cpumask_node(node);
+
 #ifdef CONFIG_SCHED_SMT
 	/*
 	 * SMT mask should be cleared whether we can claim @cpu or not. The SMT
@@ -3181,29 +3236,34 @@ static bool test_and_clear_cpu_idle(int cpu)
 	 */
 	if (sched_smt_active()) {
 		const struct cpumask *smt = cpu_smt_mask(cpu);
+		struct cpumask *idle_smt = get_idle_smtmask_node(node);
 
 		/*
 		 * If offline, @cpu is not its own sibling and
 		 * scx_pick_idle_cpu() can get caught in an infinite loop as
-		 * @cpu is never cleared from idle_masks.smt. Ensure that @cpu
-		 * is eventually cleared.
+		 * @cpu is never cleared from the idle SMT mask. Ensure that
+		 * @cpu is eventually cleared.
+		 *
+		 * NOTE: Use cpumask_intersects() and cpumask_test_cpu() to
+		 * reduce memory writes, which may help alleviate cache
+		 * coherence pressure.
 		 */
-		if (cpumask_intersects(smt, idle_masks.smt))
-			cpumask_andnot(idle_masks.smt, idle_masks.smt, smt);
-		else if (cpumask_test_cpu(cpu, idle_masks.smt))
-			__cpumask_clear_cpu(cpu, idle_masks.smt);
+		if (cpumask_intersects(smt, idle_smt))
+			cpumask_andnot(idle_smt, idle_smt, smt);
+		else if (cpumask_test_cpu(cpu, idle_smt))
+			__cpumask_clear_cpu(cpu, idle_smt);
 	}
 #endif
-	return cpumask_test_and_clear_cpu(cpu, idle_masks.cpu);
+	return cpumask_test_and_clear_cpu(cpu, idle_cpu);
 }
 
-static s32 scx_pick_idle_cpu(const struct cpumask *cpus_allowed, u64 flags)
+static s32 scx_pick_idle_cpu_from_node(int node, const struct cpumask *cpus_allowed, u64 flags)
 {
 	int cpu;
 
 retry:
 	if (sched_smt_active()) {
-		cpu = cpumask_any_and_distribute(idle_masks.smt, cpus_allowed);
+		cpu = cpumask_any_and_distribute(get_idle_smtmask_node(node), cpus_allowed);
 		if (cpu < nr_cpu_ids)
 			goto found;
 
@@ -3211,15 +3271,66 @@ static s32 scx_pick_idle_cpu(const struct cpumask *cpus_allowed, u64 flags)
 			return -EBUSY;
 	}
 
-	cpu = cpumask_any_and_distribute(idle_masks.cpu, cpus_allowed);
-	if (cpu >= nr_cpu_ids)
-		return -EBUSY;
+	cpu = cpumask_any_and_distribute(get_idle_cpumask_node(node), cpus_allowed);
+	if (cpu < nr_cpu_ids)
+		goto found;
+
+	return -EBUSY;
 
 found:
 	if (test_and_clear_cpu_idle(cpu))
 		return cpu;
 	else
 		goto retry;
+
+}
+
+static s32 scx_pick_idle_cpu(const struct cpumask *cpus_allowed, s32 prev_cpu, u64 flags)
+{
+	const struct cpumask *node_mask;
+	s32 cpu;
+
+	/*
+	 * Only node 0 is used if per-node idle cpumasks are disabled.
+	 */
+	if (!static_branch_maybe(CONFIG_NUMA, &scx_builtin_idle_per_node))
+		return scx_pick_idle_cpu_from_node(0, cpus_allowed, flags);
+
+	/*
+	 * Traverse all nodes in order of increasing distance, starting from
+	 * prev_cpu's node.
+	 */
+	rcu_read_lock();
+	for_each_numa_hop_mask(node_mask, cpu_to_node(prev_cpu)) {
+		/*
+		 * scx_pick_idle_cpu_from_node() can be expensive and redundant
+		 * if none of the CPUs in the NUMA node can be used (according
+		 * to cpus_allowed).
+		 *
+		 * Therefore, check if the NUMA node is usable in advance to
+		 * save some CPU cycles.
+		 */
+		if (!cpumask_intersects(node_mask, cpus_allowed))
+			continue;
+
+		/*
+		 * It would be nice to have a "node" iterator, instead of the
+		 * cpumask, to get rid of the cpumask_first() to determine the
+		 * node.
+		 */
+		cpu = cpumask_first(node_mask);
+		if (cpu >= nr_cpu_ids)
+			continue;
+
+		cpu = scx_pick_idle_cpu_from_node(cpu_to_node(cpu), cpus_allowed, flags);
+		if (cpu >= 0)
+			goto out_unlock;
+	}
+	cpu = -EBUSY;
+
+out_unlock:
+	rcu_read_unlock();
+	return cpu;
 }
 
 /*
@@ -3427,6 +3538,7 @@ static s32 scx_select_cpu_dfl(struct task_struct *p, s32 prev_cpu,
 {
 	const struct cpumask *llc_cpus = NULL;
 	const struct cpumask *numa_cpus = NULL;
+	int node = cpu_to_node(prev_cpu);
 	s32 cpu;
 
 	*found = false;
@@ -3484,9 +3596,9 @@ static s32 scx_select_cpu_dfl(struct task_struct *p, s32 prev_cpu,
 		 * piled up on it even if there is an idle core elsewhere on
 		 * the system.
 		 */
-		if (!cpumask_empty(idle_masks.cpu) &&
-		    !(current->flags & PF_EXITING) &&
-		    cpu_rq(cpu)->scx.local_dsq.nr == 0) {
+		if (!(current->flags & PF_EXITING) &&
+		    cpu_rq(cpu)->scx.local_dsq.nr == 0 &&
+		    !cpumask_empty(get_idle_cpumask_node(cpu_to_node(cpu)))) {
 			if (cpumask_test_cpu(cpu, p->cpus_ptr))
 				goto cpu_found;
 		}
@@ -3500,7 +3612,7 @@ static s32 scx_select_cpu_dfl(struct task_struct *p, s32 prev_cpu,
 		/*
 		 * Keep using @prev_cpu if it's part of a fully idle core.
 		 */
-		if (cpumask_test_cpu(prev_cpu, idle_masks.smt) &&
+		if (cpumask_test_cpu(prev_cpu, get_idle_smtmask_node(node)) &&
 		    test_and_clear_cpu_idle(prev_cpu)) {
 			cpu = prev_cpu;
 			goto cpu_found;
@@ -3510,7 +3622,7 @@ static s32 scx_select_cpu_dfl(struct task_struct *p, s32 prev_cpu,
 		 * Search for any fully idle core in the same LLC domain.
 		 */
 		if (llc_cpus) {
-			cpu = scx_pick_idle_cpu(llc_cpus, SCX_PICK_IDLE_CORE);
+			cpu = scx_pick_idle_cpu_from_node(node, llc_cpus, SCX_PICK_IDLE_CORE);
 			if (cpu >= 0)
 				goto cpu_found;
 		}
@@ -3519,7 +3631,7 @@ static s32 scx_select_cpu_dfl(struct task_struct *p, s32 prev_cpu,
 		 * Search for any fully idle core in the same NUMA node.
 		 */
 		if (numa_cpus) {
-			cpu = scx_pick_idle_cpu(numa_cpus, SCX_PICK_IDLE_CORE);
+			cpu = scx_pick_idle_cpu_from_node(node, numa_cpus, SCX_PICK_IDLE_CORE);
 			if (cpu >= 0)
 				goto cpu_found;
 		}
@@ -3527,7 +3639,7 @@ static s32 scx_select_cpu_dfl(struct task_struct *p, s32 prev_cpu,
 		/*
 		 * Search for any full idle core usable by the task.
 		 */
-		cpu = scx_pick_idle_cpu(p->cpus_ptr, SCX_PICK_IDLE_CORE);
+		cpu = scx_pick_idle_cpu(p->cpus_ptr, prev_cpu, SCX_PICK_IDLE_CORE);
 		if (cpu >= 0)
 			goto cpu_found;
 	}
@@ -3544,7 +3656,7 @@ static s32 scx_select_cpu_dfl(struct task_struct *p, s32 prev_cpu,
 	 * Search for any idle CPU in the same LLC domain.
 	 */
 	if (llc_cpus) {
-		cpu = scx_pick_idle_cpu(llc_cpus, 0);
+		cpu = scx_pick_idle_cpu_from_node(node, llc_cpus, 0);
 		if (cpu >= 0)
 			goto cpu_found;
 	}
@@ -3553,7 +3665,7 @@ static s32 scx_select_cpu_dfl(struct task_struct *p, s32 prev_cpu,
 	 * Search for any idle CPU in the same NUMA node.
 	 */
 	if (numa_cpus) {
-		cpu = scx_pick_idle_cpu(numa_cpus, 0);
+		cpu = scx_pick_idle_cpu_from_node(node, numa_cpus, 0);
 		if (cpu >= 0)
 			goto cpu_found;
 	}
@@ -3561,7 +3673,7 @@ static s32 scx_select_cpu_dfl(struct task_struct *p, s32 prev_cpu,
 	/*
 	 * Search for any idle CPU usable by the task.
 	 */
-	cpu = scx_pick_idle_cpu(p->cpus_ptr, 0);
+	cpu = scx_pick_idle_cpu(p->cpus_ptr, prev_cpu, 0);
 	if (cpu >= 0)
 		goto cpu_found;
 
@@ -3643,17 +3755,27 @@ static void set_cpus_allowed_scx(struct task_struct *p,
 
 static void reset_idle_masks(void)
 {
+	int node;
+
 	/*
 	 * Consider all online cpus idle. Should converge to the actual state
 	 * quickly.
 	 */
-	cpumask_copy(idle_masks.cpu, cpu_online_mask);
-	cpumask_copy(idle_masks.smt, cpu_online_mask);
+	for_each_node_state(node, N_POSSIBLE) {
+		const struct cpumask *node_mask = cpumask_of_node(node);
+		struct cpumask *idle_cpu = get_idle_cpumask_node(node);
+		struct cpumask *idle_smt = get_idle_smtmask_node(node);
+
+		cpumask_and(idle_cpu, cpu_online_mask, node_mask);
+		cpumask_copy(idle_smt, idle_cpu);
+	}
 }
 
 void __scx_update_idle(struct rq *rq, bool idle)
 {
 	int cpu = cpu_of(rq);
+	int node = cpu_to_node(cpu);
+	struct cpumask *idle_cpu = get_idle_cpumask_node(node);
 
 	if (SCX_HAS_OP(update_idle) && !scx_rq_bypassing(rq)) {
 		SCX_CALL_OP(SCX_KF_REST, update_idle, cpu_of(rq), idle);
@@ -3661,27 +3783,25 @@ void __scx_update_idle(struct rq *rq, bool idle)
 			return;
 	}
 
-	if (idle)
-		cpumask_set_cpu(cpu, idle_masks.cpu);
-	else
-		cpumask_clear_cpu(cpu, idle_masks.cpu);
+	assign_cpu(cpu, idle_cpu, idle);
 
 #ifdef CONFIG_SCHED_SMT
 	if (sched_smt_active()) {
 		const struct cpumask *smt = cpu_smt_mask(cpu);
+		struct cpumask *idle_smt = get_idle_smtmask_node(node);
 
 		if (idle) {
 			/*
-			 * idle_masks.smt handling is racy but that's fine as
-			 * it's only for optimization and self-correcting.
+			 * idle_smt handling is racy but that's fine as it's
+			 * only for optimization and self-correcting.
 			 */
 			for_each_cpu(cpu, smt) {
-				if (!cpumask_test_cpu(cpu, idle_masks.cpu))
+				if (!cpumask_test_cpu(cpu, idle_cpu))
 					return;
 			}
-			cpumask_or(idle_masks.smt, idle_masks.smt, smt);
+			cpumask_or(idle_smt, idle_smt, smt);
 		} else {
-			cpumask_andnot(idle_masks.smt, idle_masks.smt, smt);
+			cpumask_andnot(idle_smt, idle_smt, smt);
 		}
 	}
 #endif
@@ -3729,7 +3849,10 @@ static void rq_offline_scx(struct rq *rq)
 #else	/* CONFIG_SMP */
 
 static bool test_and_clear_cpu_idle(int cpu) { return false; }
-static s32 scx_pick_idle_cpu(const struct cpumask *cpus_allowed, u64 flags) { return -EBUSY; }
+static s32 scx_pick_idle_cpu(const struct cpumask *cpus_allowed, s32 prev_cpu, u64 flags)
+{
+	return -EBUSY;
+}
 static void reset_idle_masks(void) {}
 
 #endif	/* CONFIG_SMP */
@@ -6262,8 +6385,7 @@ void __init init_sched_ext_class(void)
 
 	BUG_ON(rhashtable_init(&dsq_hash, &dsq_hash_params));
 #ifdef CONFIG_SMP
-	BUG_ON(!alloc_cpumask_var(&idle_masks.cpu, GFP_KERNEL));
-	BUG_ON(!alloc_cpumask_var(&idle_masks.smt, GFP_KERNEL));
+	idle_masks_init();
 #endif
 	scx_kick_cpus_pnt_seqs =
 		__alloc_percpu(sizeof(scx_kick_cpus_pnt_seqs[0]) * nr_cpu_ids,
@@ -7409,7 +7531,7 @@ __bpf_kfunc void scx_bpf_put_cpumask(const struct cpumask *cpumask)
 
 /**
  * scx_bpf_get_idle_cpumask - Get a referenced kptr to the idle-tracking
- * per-CPU cpumask.
+ * per-CPU cpumask of the current NUMA node.
  *
  * Returns NULL if idle tracking is not enabled, or running on a UP kernel.
  */
@@ -7420,17 +7542,13 @@ __bpf_kfunc const struct cpumask *scx_bpf_get_idle_cpumask(void)
 		return cpu_none_mask;
 	}
 
-#ifdef CONFIG_SMP
-	return idle_masks.cpu;
-#else
-	return cpu_none_mask;
-#endif
+	return get_curr_idle_cpumask();
 }
 
 /**
  * scx_bpf_get_idle_smtmask - Get a referenced kptr to the idle-tracking,
- * per-physical-core cpumask. Can be used to determine if an entire physical
- * core is free.
+ * per-physical-core cpumask of the current NUMA node. Can be used to determine
+ * if an entire physical core is free.
  *
  * Returns NULL if idle tracking is not enabled, or running on a UP kernel.
  */
@@ -7441,14 +7559,7 @@ __bpf_kfunc const struct cpumask *scx_bpf_get_idle_smtmask(void)
 		return cpu_none_mask;
 	}
 
-#ifdef CONFIG_SMP
-	if (sched_smt_active())
-		return idle_masks.smt;
-	else
-		return idle_masks.cpu;
-#else
-	return cpu_none_mask;
-#endif
+	return get_curr_idle_smtmask();
 }
 
 /**
@@ -7515,7 +7626,7 @@ __bpf_kfunc s32 scx_bpf_pick_idle_cpu(const struct cpumask *cpus_allowed,
 		return -EBUSY;
 	}
 
-	return scx_pick_idle_cpu(cpus_allowed, flags);
+	return scx_pick_idle_cpu(cpus_allowed, smp_processor_id(), flags);
 }
 
 /**
@@ -7538,7 +7649,7 @@ __bpf_kfunc s32 scx_bpf_pick_any_cpu(const struct cpumask *cpus_allowed,
 	s32 cpu;
 
 	if (static_branch_likely(&scx_builtin_idle_enabled)) {
-		cpu = scx_pick_idle_cpu(cpus_allowed, flags);
+		cpu = scx_pick_idle_cpu(cpus_allowed, smp_processor_id(), flags);
 		if (cpu >= 0)
 			return cpu;
 	}
Signature · 1 line
2.47.1
author reviewer
#3 lore raw

PATCH 3/6 sched_ext: Get rid of the scx_selcpu_topo_numa logic

In reply to Andrea Righi

With the introduction of separate per-NUMA node cpumasks, we automatically track idle CPUs within each NUMA node.

This makes the special logic for determining idle CPUs in each NUMA node redundant and unnecessary, so we can get rid of it.

---
 kernel/sched/ext.c | 77 +++-------------------------------------------
 1 file changed, 5 insertions(+), 72 deletions(-)
kernel/sched/ext.c +5 -72
index 0b9b6627515b..a07a76c08f81 100644
@@ -893,7 +893,6 @@ static DEFINE_STATIC_KEY_FALSE(scx_builtin_idle_enabled);
 
 #ifdef CONFIG_SMP
 static DEFINE_STATIC_KEY_FALSE(scx_selcpu_topo_llc);
-static DEFINE_STATIC_KEY_FALSE(scx_selcpu_topo_numa);
 #endif
 
 static struct static_key_false scx_has_op[SCX_OPI_END] =
@@ -3382,25 +3381,6 @@ static unsigned int numa_weight(s32 cpu)
 	return sg->group_weight;
 }
 
-/*
- * Return the cpumask representing the NUMA domain of @cpu (or NULL if the NUMA
- * domain is not defined).
- */
-static struct cpumask *numa_span(s32 cpu)
-{
-	struct sched_domain *sd;
-	struct sched_group *sg;
-
-	sd = rcu_dereference(per_cpu(sd_numa, cpu));
-	if (!sd)
-		return NULL;
-	sg = sd->groups;
-	if (!sg)
-		return NULL;
-
-	return sched_group_span(sg);
-}
-
 /*
  * Return true if the LLC domains do not perfectly overlap with the NUMA
  * domains, false otherwise.
@@ -3452,7 +3432,7 @@ static bool llc_numa_mismatch(void)
  */
 static void update_selcpu_topology(void)
 {
-	bool enable_llc = false, enable_numa = false;
+	bool enable_llc = false;
 	unsigned int nr_cpus;
 	s32 cpu = cpumask_first(cpu_online_mask);
 
@@ -3469,43 +3449,20 @@ static void update_selcpu_topology(void)
 	rcu_read_lock();
 	nr_cpus = llc_weight(cpu);
 	if (nr_cpus > 0) {
-		if (nr_cpus < num_online_cpus())
+		if ((nr_cpus < num_online_cpus()) && llc_numa_mismatch())
 			enable_llc = true;
 		pr_debug("sched_ext: LLC=%*pb weight=%u\n",
 			 cpumask_pr_args(llc_span(cpu)), llc_weight(cpu));
 	}
-
-	/*
-	 * Enable NUMA optimization only when there are multiple NUMA domains
-	 * among the online CPUs and the NUMA domains don't perfectly overlaps
-	 * with the LLC domains.
-	 *
-	 * If all CPUs belong to the same NUMA node and the same LLC domain,
-	 * enabling both NUMA and LLC optimizations is unnecessary, as checking
-	 * for an idle CPU in the same domain twice is redundant.
-	 */
-	nr_cpus = numa_weight(cpu);
-	if (nr_cpus > 0) {
-		if (nr_cpus < num_online_cpus() && llc_numa_mismatch())
-			enable_numa = true;
-		pr_debug("sched_ext: NUMA=%*pb weight=%u\n",
-			 cpumask_pr_args(numa_span(cpu)), numa_weight(cpu));
-	}
 	rcu_read_unlock();
 
 	pr_debug("sched_ext: LLC idle selection %s\n",
 		 enable_llc ? "enabled" : "disabled");
-	pr_debug("sched_ext: NUMA idle selection %s\n",
-		 enable_numa ? "enabled" : "disabled");
 
 	if (enable_llc)
 		static_branch_enable_cpuslocked(&scx_selcpu_topo_llc);
 	else
 		static_branch_disable_cpuslocked(&scx_selcpu_topo_llc);
-	if (enable_numa)
-		static_branch_enable_cpuslocked(&scx_selcpu_topo_numa);
-	else
-		static_branch_disable_cpuslocked(&scx_selcpu_topo_numa);
 }
 
 /*
@@ -3526,9 +3483,8 @@ static void update_selcpu_topology(void)
  * 4. Pick a CPU within the same NUMA node, if enabled:
  *   - choose a CPU from the same NUMA node to reduce memory access latency.
  *
- * Step 3 and 4 are performed only if the system has, respectively, multiple
- * LLC domains / multiple NUMA nodes (see scx_selcpu_topo_llc and
- * scx_selcpu_topo_numa).
+ * Step 3 is performed only if the system has multiple LLC domains that are not
+ * perfectly overlapping with the NUMA domains (see scx_selcpu_topo_llc).
  *
  * NOTE: tasks that can only run on 1 CPU are excluded by this logic, because
  * we never call ops.select_cpu() for them, see select_task_rq().
@@ -3537,7 +3493,6 @@ static s32 scx_select_cpu_dfl(struct task_struct *p, s32 prev_cpu,
 			      u64 wake_flags, bool *found)
 {
 	const struct cpumask *llc_cpus = NULL;
-	const struct cpumask *numa_cpus = NULL;
 	int node = cpu_to_node(prev_cpu);
 	s32 cpu;
 
@@ -3559,13 +3514,9 @@ static s32 scx_select_cpu_dfl(struct task_struct *p, s32 prev_cpu,
 	 * CPU affinity), the task will simply use the flat scheduling domain
 	 * defined by user-space.
 	 */
-	if (p->nr_cpus_allowed >= num_possible_cpus()) {
-		if (static_branch_maybe(CONFIG_NUMA, &scx_selcpu_topo_numa))
-			numa_cpus = numa_span(prev_cpu);
-
+	if (p->nr_cpus_allowed >= num_possible_cpus())
 		if (static_branch_maybe(CONFIG_SCHED_MC, &scx_selcpu_topo_llc))
 			llc_cpus = llc_span(prev_cpu);
-	}
 
 	/*
 	 * If WAKE_SYNC, try to migrate the wakee to the waker's CPU.
@@ -3627,15 +3578,6 @@ static s32 scx_select_cpu_dfl(struct task_struct *p, s32 prev_cpu,
 				goto cpu_found;
 		}
 
-		/*
-		 * Search for any fully idle core in the same NUMA node.
-		 */
-		if (numa_cpus) {
-			cpu = scx_pick_idle_cpu_from_node(node, numa_cpus, SCX_PICK_IDLE_CORE);
-			if (cpu >= 0)
-				goto cpu_found;
-		}
-
 		/*
 		 * Search for any full idle core usable by the task.
 		 */
@@ -3661,15 +3603,6 @@ static s32 scx_select_cpu_dfl(struct task_struct *p, s32 prev_cpu,
 			goto cpu_found;
 	}
 
-	/*
-	 * Search for any idle CPU in the same NUMA node.
-	 */
-	if (numa_cpus) {
-		cpu = scx_pick_idle_cpu_from_node(node, numa_cpus, 0);
-		if (cpu >= 0)
-			goto cpu_found;
-	}
-
 	/*
 	 * Search for any idle CPU usable by the task.
 	 */
Signature · 1 line
2.47.1
author reviewer
#4 lore raw

PATCH 4/6 sched_ext: Use SCX_OPS_NODE_BUILTIN_IDLE

In reply to Andrea Righi

Use the new SCX_OPS_NODE_BUILTIN_IDLE flat to allow each scx scheduler to choose between a flat idle cpumask and multiple per-node cpumasks.

---
 kernel/sched/ext.c | 49 ++++++++++++++++++++++++++++++++++++++++------
 1 file changed, 43 insertions(+), 6 deletions(-)
kernel/sched/ext.c +43 -6
index a07a76c08f81..d0d57323bcfc 100644
@@ -893,6 +893,7 @@ static DEFINE_STATIC_KEY_FALSE(scx_builtin_idle_enabled);
 
 #ifdef CONFIG_SMP
 static DEFINE_STATIC_KEY_FALSE(scx_selcpu_topo_llc);
+static DEFINE_STATIC_KEY_FALSE(scx_builtin_idle_per_node);
 #endif
 
 static struct static_key_false scx_has_op[SCX_OPI_END] =
@@ -936,18 +937,32 @@ static struct delayed_work scx_watchdog_work;
 #define CL_ALIGNED_IF_ONSTACK __cacheline_aligned_in_smp
 #endif
 
-static struct {
+struct idle_cpumask {
 	cpumask_var_t cpu;
 	cpumask_var_t smt;
-} **idle_masks CL_ALIGNED_IF_ONSTACK;
+};
+
+/*
+ * cpumasks to track idle CPUs within each NUMA node.
+ *
+ * If SCX_OPS_BUILTIN_IDLE_PER_NODE is not specified, a single flat cpumask
+ * from node 0 is used to track all idle CPUs system-wide.
+ */
+static struct idle_cpumask **idle_masks CL_ALIGNED_IF_ONSTACK;
 
 static struct cpumask *get_idle_cpumask_node(int node)
 {
+	if (!static_branch_maybe(CONFIG_NUMA, &scx_builtin_idle_per_node))
+		return idle_masks[0]->cpu;
+
 	return idle_masks[node]->cpu;
 }
 
 static struct cpumask *get_idle_smtmask_node(int node)
 {
+	if (!static_branch_maybe(CONFIG_NUMA, &scx_builtin_idle_per_node))
+		return idle_masks[0]->smt;
+
 	return idle_masks[node]->smt;
 }
 
@@ -3430,7 +3445,7 @@ static bool llc_numa_mismatch(void)
  * CPU belongs to a single LLC domain, and that each LLC domain is entirely
  * contained within a single NUMA node.
  */
-static void update_selcpu_topology(void)
+static void update_selcpu_topology(struct sched_ext_ops *ops)
 {
 	bool enable_llc = false;
 	unsigned int nr_cpus;
@@ -3449,8 +3464,16 @@ static void update_selcpu_topology(void)
 	rcu_read_lock();
 	nr_cpus = llc_weight(cpu);
 	if (nr_cpus > 0) {
-		if ((nr_cpus < num_online_cpus()) && llc_numa_mismatch())
+		if (nr_cpus < num_online_cpus())
 			enable_llc = true;
+		/*
+		 * No need to enable LLC optimization if the LLC domains are
+		 * perfectly overlapping with the NUMA domains when per-node
+		 * cpumasks are enabled.
+		 */
+		if ((ops->flags & SCX_OPS_BUILTIN_IDLE_PER_NODE) &&
+		    !llc_numa_mismatch())
+			enable_llc = false;
 		pr_debug("sched_ext: LLC=%*pb weight=%u\n",
 			 cpumask_pr_args(llc_span(cpu)), llc_weight(cpu));
 	}
@@ -3463,6 +3486,14 @@ static void update_selcpu_topology(void)
 		static_branch_enable_cpuslocked(&scx_selcpu_topo_llc);
 	else
 		static_branch_disable_cpuslocked(&scx_selcpu_topo_llc);
+
+	/*
+	 * Check if we need to enable per-node cpumasks.
+	 */
+	if (ops->flags & SCX_OPS_BUILTIN_IDLE_PER_NODE)
+		static_branch_enable_cpuslocked(&scx_builtin_idle_per_node);
+	else
+		static_branch_disable_cpuslocked(&scx_builtin_idle_per_node);
 }
 
 /*
@@ -3690,6 +3721,12 @@ static void reset_idle_masks(void)
 {
 	int node;
 
+	if (!static_branch_maybe(CONFIG_NUMA, &scx_builtin_idle_per_node)) {
+		cpumask_copy(get_idle_cpumask_node(0), cpu_online_mask);
+		cpumask_copy(get_idle_smtmask_node(0), cpu_online_mask);
+		return;
+	}
+
 	/*
 	 * Consider all online cpus idle. Should converge to the actual state
 	 * quickly.
@@ -3747,7 +3784,7 @@ static void handle_hotplug(struct rq *rq, bool online)
 	atomic_long_inc(&scx_hotplug_seq);
 
 	if (scx_enabled())
-		update_selcpu_topology();
+		update_selcpu_topology(&scx_ops);
 
 	if (online && SCX_HAS_OP(cpu_online))
 		SCX_CALL_OP(SCX_KF_UNLOCKED, cpu_online, cpu);
@@ -5625,7 +5662,7 @@ static int scx_ops_enable(struct sched_ext_ops *ops, struct bpf_link *link)
 
 	check_hotplug_seq(ops);
 #ifdef CONFIG_SMP
-	update_selcpu_topology();
+	update_selcpu_topology(ops);
 #endif
 	cpus_read_unlock();
 
Signature · 1 line
2.47.1
author reviewer
#5 lore raw

PATCH 5/6 sched_ext: Introduce NUMA aware idle cpu kfunc helpers

In reply to Andrea Righi

Add the following kfunc's to provide scx schedulers direct access to per-node idle cpumasks information:

const struct cpumask *scx_bpf_get_idle_cpumask_node(int node) const struct cpumask *scx_bpf_get_idle_smtmask_node(int node) s32 scx_bpf_pick_idle_cpu_node(int node,
const cpumask_t *cpus_allowed, u64 flags)
int scx_bpf_cpu_to_node(s32 cpu)

---
 kernel/sched/ext.c                       | 96 +++++++++++++++++++++++-
 tools/sched_ext/include/scx/common.bpf.h |  4 +
 tools/sched_ext/include/scx/compat.bpf.h | 19 +++++
 3 files changed, 117 insertions(+), 2 deletions(-)
kernel/sched/ext.c +94 -2
index d0d57323bcfc..ea7cc481782c 100644
@@ -433,6 +433,7 @@ struct sched_ext_ops {
 	 * - scx_bpf_select_cpu_dfl()
 	 * - scx_bpf_test_and_clear_cpu_idle()
 	 * - scx_bpf_pick_idle_cpu()
+	 * - scx_bpf_pick_idle_cpu_node()
 	 *
 	 * The user also must implement ops.select_cpu() as the default
 	 * implementation relies on scx_bpf_select_cpu_dfl().
@@ -955,6 +956,8 @@ static struct cpumask *get_idle_cpumask_node(int node)
 	if (!static_branch_maybe(CONFIG_NUMA, &scx_builtin_idle_per_node))
 		return idle_masks[0]->cpu;
 
+	if (node < 0 || node >= num_possible_nodes())
+		return NULL;
 	return idle_masks[node]->cpu;
 }
 
@@ -963,6 +966,8 @@ static struct cpumask *get_idle_smtmask_node(int node)
 	if (!static_branch_maybe(CONFIG_NUMA, &scx_builtin_idle_per_node))
 		return idle_masks[0]->smt;
 
+	if (node < 0 || node >= num_possible_nodes())
+		return NULL;
 	return idle_masks[node]->smt;
 }
 
@@ -7469,6 +7474,16 @@ __bpf_kfunc u32 scx_bpf_nr_cpu_ids(void)
 	return nr_cpu_ids;
 }
 
+/**
+ * scx_bpf_cpu_to_node - Return the NUMA node the given @cpu belongs to
+ */
+__bpf_kfunc int scx_bpf_cpu_to_node(s32 cpu)
+{
+	if (cpu < 0 || cpu >= nr_cpu_ids)
+		return -EINVAL;
+	return cpu_to_node(cpu);
+}
+
 /**
  * scx_bpf_get_possible_cpumask - Get a referenced kptr to cpu_possible_mask
  */
@@ -7499,11 +7514,32 @@ __bpf_kfunc void scx_bpf_put_cpumask(const struct cpumask *cpumask)
 	 */
 }
 
+/**
+ * scx_bpf_get_idle_cpumask_node - Get a referenced kptr to the idle-tracking
+ * per-CPU cpumask of a target NUMA node.
+ *
+ * Returns an empty cpumask if idle tracking is not enabled, if @node is not
+ * valid, or running on a UP kernel.
+ */
+__bpf_kfunc const struct cpumask *scx_bpf_get_idle_cpumask_node(int node)
+{
+	if (!static_branch_likely(&scx_builtin_idle_enabled)) {
+		scx_ops_error("built-in idle tracking is disabled");
+		return cpu_none_mask;
+	}
+	if (!static_branch_likely(&scx_builtin_idle_per_node)) {
+		scx_ops_error("per-node idle tracking is disabled");
+		return cpu_none_mask;
+	}
+
+	return get_idle_cpumask_node(node) ? : cpu_none_mask;
+}
 /**
  * scx_bpf_get_idle_cpumask - Get a referenced kptr to the idle-tracking
  * per-CPU cpumask of the current NUMA node.
  *
- * Returns NULL if idle tracking is not enabled, or running on a UP kernel.
+ * Returns an emtpy cpumask if idle tracking is not enabled, or running on a UP
+ * kernel.
  */
 __bpf_kfunc const struct cpumask *scx_bpf_get_idle_cpumask(void)
 {
@@ -7515,12 +7551,35 @@ __bpf_kfunc const struct cpumask *scx_bpf_get_idle_cpumask(void)
 	return get_curr_idle_cpumask();
 }
 
+/**
+ * scx_bpf_get_idle_smtmask_node - Get a referenced kptr to the idle-tracking,
+ * per-physical-core cpumask of a target NUMA node. Can be used to determine
+ * if an entire physical core is free.
+ *
+ * Returns an empty cpumask if idle tracking is not enabled, if @node is not
+ * valid, or running on a UP kernel.
+ */
+__bpf_kfunc const struct cpumask *scx_bpf_get_idle_smtmask_node(int node)
+{
+	if (!static_branch_likely(&scx_builtin_idle_enabled)) {
+		scx_ops_error("built-in idle tracking is disabled");
+		return cpu_none_mask;
+	}
+	if (!static_branch_likely(&scx_builtin_idle_per_node)) {
+		scx_ops_error("per-node idle tracking is disabled");
+		return cpu_none_mask;
+	}
+
+	return get_idle_smtmask_node(node) ? : cpu_none_mask;
+}
+
 /**
  * scx_bpf_get_idle_smtmask - Get a referenced kptr to the idle-tracking,
  * per-physical-core cpumask of the current NUMA node. Can be used to determine
  * if an entire physical core is free.
  *
- * Returns NULL if idle tracking is not enabled, or running on a UP kernel.
+ * Returns an empty cumask if idle tracking is not enabled, or running on a UP
+ * kernel.
  */
 __bpf_kfunc const struct cpumask *scx_bpf_get_idle_smtmask(void)
 {
@@ -7569,6 +7628,35 @@ __bpf_kfunc bool scx_bpf_test_and_clear_cpu_idle(s32 cpu)
 		return false;
 }
 
+/**
+ * scx_bpf_pick_idle_cpu_node - Pick and claim an idle cpu from a NUMA node
+ * @node: target NUMA node
+ * @cpus_allowed: Allowed cpumask
+ * @flags: %SCX_PICK_IDLE_CPU_* flags
+ *
+ * Pick and claim an idle cpu in @cpus_allowed from the NUMA node @node.
+ * Returns the picked idle cpu number on success. -%EBUSY if no matching cpu
+ * was found.
+ *
+ * Unavailable if ops.update_idle() is implemented and
+ * %SCX_OPS_KEEP_BUILTIN_IDLE is not set or if %SCX_OPS_KEEP_BUILTIN_IDLE is
+ * not set.
+ */
+__bpf_kfunc s32 scx_bpf_pick_idle_cpu_node(int node, const struct cpumask *cpus_allowed,
+				      u64 flags)
+{
+	if (!static_branch_likely(&scx_builtin_idle_enabled)) {
+		scx_ops_error("built-in idle tracking is disabled");
+		return -EBUSY;
+	}
+	if (!static_branch_likely(&scx_builtin_idle_per_node)) {
+		scx_ops_error("per-node idle tracking is disabled");
+		return -EBUSY;
+	}
+
+	return scx_pick_idle_cpu_from_node(node, cpus_allowed, flags);
+}
+
 /**
  * scx_bpf_pick_idle_cpu - Pick and claim an idle cpu
  * @cpus_allowed: Allowed cpumask
@@ -7705,14 +7793,18 @@ BTF_ID_FLAGS(func, scx_bpf_cpuperf_cap)
 BTF_ID_FLAGS(func, scx_bpf_cpuperf_cur)
 BTF_ID_FLAGS(func, scx_bpf_cpuperf_set)
 BTF_ID_FLAGS(func, scx_bpf_nr_cpu_ids)
+BTF_ID_FLAGS(func, scx_bpf_cpu_to_node)
 BTF_ID_FLAGS(func, scx_bpf_get_possible_cpumask, KF_ACQUIRE)
 BTF_ID_FLAGS(func, scx_bpf_get_online_cpumask, KF_ACQUIRE)
 BTF_ID_FLAGS(func, scx_bpf_put_cpumask, KF_RELEASE)
 BTF_ID_FLAGS(func, scx_bpf_get_idle_cpumask, KF_ACQUIRE)
+BTF_ID_FLAGS(func, scx_bpf_get_idle_cpumask_node, KF_ACQUIRE)
 BTF_ID_FLAGS(func, scx_bpf_get_idle_smtmask, KF_ACQUIRE)
+BTF_ID_FLAGS(func, scx_bpf_get_idle_smtmask_node, KF_ACQUIRE)
 BTF_ID_FLAGS(func, scx_bpf_put_idle_cpumask, KF_RELEASE)
 BTF_ID_FLAGS(func, scx_bpf_test_and_clear_cpu_idle)
 BTF_ID_FLAGS(func, scx_bpf_pick_idle_cpu, KF_RCU)
+BTF_ID_FLAGS(func, scx_bpf_pick_idle_cpu_node, KF_RCU)
 BTF_ID_FLAGS(func, scx_bpf_pick_any_cpu, KF_RCU)
 BTF_ID_FLAGS(func, scx_bpf_task_running, KF_RCU)
 BTF_ID_FLAGS(func, scx_bpf_task_cpu, KF_RCU)
tools/sched_ext/include/scx/common.bpf.h +4 -0
index 625f5b046776..9bbf6d5083b5 100644
@@ -59,14 +59,18 @@ u32 scx_bpf_cpuperf_cap(s32 cpu) __ksym __weak;
 u32 scx_bpf_cpuperf_cur(s32 cpu) __ksym __weak;
 void scx_bpf_cpuperf_set(s32 cpu, u32 perf) __ksym __weak;
 u32 scx_bpf_nr_cpu_ids(void) __ksym __weak;
+int scx_bpf_cpu_to_node(s32 cpu) __ksym __weak;
 const struct cpumask *scx_bpf_get_possible_cpumask(void) __ksym __weak;
 const struct cpumask *scx_bpf_get_online_cpumask(void) __ksym __weak;
 void scx_bpf_put_cpumask(const struct cpumask *cpumask) __ksym __weak;
 const struct cpumask *scx_bpf_get_idle_cpumask(void) __ksym;
+const struct cpumask *scx_bpf_get_idle_cpumask_node(int node) __ksym __weak;
 const struct cpumask *scx_bpf_get_idle_smtmask(void) __ksym;
+const struct cpumask *scx_bpf_get_idle_smtmask_node(int node) __ksym __weak;
 void scx_bpf_put_idle_cpumask(const struct cpumask *cpumask) __ksym;
 bool scx_bpf_test_and_clear_cpu_idle(s32 cpu) __ksym;
 s32 scx_bpf_pick_idle_cpu(const cpumask_t *cpus_allowed, u64 flags) __ksym;
+s32 scx_bpf_pick_idle_cpu_node(int node, const cpumask_t *cpus_allowed, u64 flags) __ksym __weak;
 s32 scx_bpf_pick_any_cpu(const cpumask_t *cpus_allowed, u64 flags) __ksym;
 bool scx_bpf_task_running(const struct task_struct *p) __ksym;
 s32 scx_bpf_task_cpu(const struct task_struct *p) __ksym;
tools/sched_ext/include/scx/compat.bpf.h +19 -0
index d56520100a26..587650490743 100644
@@ -125,6 +125,25 @@ bool scx_bpf_dispatch_vtime_from_dsq___compat(struct bpf_iter_scx_dsq *it__iter,
 	false;									\
 })
 
+#define __COMPAT_scx_bpf_cpu_to_node(cpu)                                       \
+	(bpf_ksym_exists(scx_bpf_cpu_to_node) ?                                 \
+	 scx_bpf_cpu_to_node(cpu) : 0)
+
+#define __COMPAT_scx_bpf_get_idle_cpumask_node(node)                            \
+	(bpf_ksym_exists(scx_bpf_get_idle_cpumask_node) ?                       \
+	 scx_bpf_get_idle_cpumask_node(node) :                                  \
+	 scx_bpf_get_idle_cpumask())                                            \
+
+#define __COMPAT_scx_bpf_get_idle_smtmask_node(node)                            \
+	(bpf_ksym_exists(scx_bpf_get_idle_smtmask_node) ?                       \
+	 scx_bpf_get_idle_smtmask_node(node) :                                  \
+	 scx_bpf_get_idle_smtmask())                                            \
+
+#define __COMPAT_scx_bpf_pick_idle_cpu_node(node, cpus_allowed, flags)          \
+	(bpf_ksym_exists(scx_bpf_pick_idle_cpu_node) ?                          \
+	 scx_bpf_pick_idle_cpu_node(node, cpus_allowed, flags) :                \
+	 scx_bpf_pick_idle_cpu(cpus_allowed, flags))
+
 /*
  * Define sched_ext_ops. This may be expanded to define multiple variants for
  * backward compatibility. See compat.h::SCX_OPS_LOAD/ATTACH().
Signature · 1 line
2.47.1
author reviewer
#6 lore raw

PATCH 6/6 sched_ext: Move built-in idle CPU selection policy to a separate file

In reply to Andrea Righi

No functional change, Only rearrange the code to move the built-in CPU idle selection logic to a separate file.

---
 MAINTAINERS             |   1 +
 kernel/sched/ext.c      | 876 +---------------------------------------
 kernel/sched/ext_idle.c | 845 ++++++++++++++++++++++++++++++++++++++
 3 files changed, 865 insertions(+), 857 deletions(-)
 create mode 100644 kernel/sched/ext_idle.c
MAINTAINERS +1 -0
index 1e930c7a58b1..02960d1b9ee9 100644
@@ -20909,6 +20909,7 @@ T:	git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext.git
 F:	include/linux/sched/ext.h
 F:	kernel/sched/ext.h
 F:	kernel/sched/ext.c
+F:	kernel/sched/ext_idle.c
 F:	tools/sched_ext/
 F:	tools/testing/selftests/sched_ext
 
kernel/sched/ext.c +19 -857

957 lines not on this page · raw message

kernel/sched/ext_idle.c new file +845 -0

845 lines not on this page · raw message

Signature · 1 line
2.47.1
author reviewer
#7 lore raw
In reply to Andrea Righi

On Mon, Dec 09, 2024 at 11:56:24PM +0100, Andrea Righi wrote:
...

Quoted 5 lines

ChangeLog v5 -> v6:
- refactor patch set to introduce SCX_OPS_NODE_BUILTIN_IDLE before the per-node cpumasks
- move idle CPU selection policy to a separate file (ext_idle.c) (no functional change, just some code shuffling)

It seems that I totally missed some comments from Yury, I'll address them in a new version. Apologize for the noise.

-Andrea