[PATCH 3/6] sched_ext: idle: Accept an arbitrary cpumask in scx_select_cpu_dfl()

[Andrea Righi]

Many scx schedulers implement their own hard or soft-affinity rules
to support topology characteristics, such as heterogeneous architectures
(e.g., big.LITTLE, P-cores/E-cores), or to categorize tasks based on
specific properties (e.g., running certain tasks only in a subset of
CPUs).

Currently, there is no mechanism that allows to use the built-in idle
CPU selection policy to an arbitrary subset of CPUs. As a result,
schedulers often implement their own idle CPU selection policies, which
are typically similar to one another, leading to a lot of code
duplication.

To address this, modify scx_select_cpu_dfl() to accept an arbitrary
cpumask, that can be used by the BPF schedulers to apply the existent
built-in idle CPU selection policy to a subset of allowed CPUs.

With this concept the idle CPU selection policy becomes the following:
 - always prioritize CPUs from fully idle SMT cores (if SMT is enabled),
 - select the same CPU if it's idle and in the allowed CPUs,
 - select an idle CPU within the same LLC, if the LLC cpumask is a
   subset of the allowed CPUs,
 - select an idle CPU within the same node, if the node cpumask is a
   subset of the allowed CPUs,
 - select an idle CPU within the allowed CPUs.

This functionality will be exposed through a dedicated kfunc in a
separate patch.

Signed-off-by: Andrea Righi <arighi@xxxxxxxxxx>
---
 kernel/sched/ext_idle.c | 96 ++++++++++++++++++++++++++++++-----------
 1 file changed, 70 insertions(+), 26 deletions(-)

diff --git a/kernel/sched/ext_idle.c b/kernel/sched/ext_idle.c
index a90d85bce1ccb..a9755434e88b7 100644
--- a/kernel/sched/ext_idle.c
+++ b/kernel/sched/ext_idle.c
@@ -49,6 +49,7 @@ static struct scx_idle_cpus **scx_idle_node_masks;
 /*
  * Local per-CPU cpumasks (used to generate temporary idle cpumasks).
  */
+static DEFINE_PER_CPU(cpumask_var_t, local_idle_cpumask);
 static DEFINE_PER_CPU(cpumask_var_t, local_llc_idle_cpumask);
 static DEFINE_PER_CPU(cpumask_var_t, local_numa_idle_cpumask);
 
@@ -397,15 +398,18 @@ void scx_idle_update_selcpu_topology(struct sched_ext_ops *ops)
 		static_branch_disable_cpuslocked(&scx_selcpu_topo_numa);
 }
 
-static inline bool task_allowed_all_cpus(const struct task_struct *p)
+/*
+ * Return true if @p can run on all possible CPUs, false otherwise.
+ */
+static inline bool task_affinity_all(const struct task_struct *p)
 {
 	return p->nr_cpus_allowed >= num_possible_cpus();
 }
 
 /*
  * Return the subset of @cpus that task @p can use, according to
- * @cpus_allowed, or NULL if none of the CPUs in the @cpus cpumask can be
- * used.
+ * @cpus_allowed, or NULL if none of the CPUs in the target cpumask @cpus
+ * can be used.
  */
 static const struct cpumask *task_cpumask(const struct task_struct *p,
 					  const struct cpumask *cpus_allowed,
@@ -414,14 +418,20 @@ static const struct cpumask *task_cpumask(const struct task_struct *p,
 {
 	/*
 	 * If the task is allowed to run on all CPUs, simply use the
-	 * architecture's cpumask directly. Otherwise, compute the
-	 * intersection of the architecture's cpumask and the task's
-	 * allowed cpumask.
+	 * target cpumask directly (@cpus). Otherwise, compute the
+	 * intersection of the target cpumask and the task's allowed
+	 * cpumask.
 	 */
-	if (!cpus || task_allowed_all_cpus(p) || cpumask_subset(cpus, cpus_allowed))
+	if (!cpus || ((cpus_allowed == p->cpus_ptr) && task_affinity_all(p)) ||
+	    cpumask_subset(cpus, cpus_allowed))
 		return cpus;
 
-	if (cpumask_and(local_cpus, cpus, cpus_allowed))
+	/*
+	 * Compute the intersection and return NULL if the result is empty
+	 * or if it perfectly overlaps with the subset of allowed CPUs.
+	 */
+	if (cpumask_and(local_cpus, cpus, cpus_allowed) &&
+	    !cpumask_equal(local_cpus, cpus_allowed))
 		return local_cpus;
 
 	return NULL;
@@ -439,13 +449,15 @@ static const struct cpumask *task_cpumask(const struct task_struct *p,
  *     branch prediction optimizations.
  *
  * 3. Pick a CPU within the same LLC (Last-Level Cache):
- *   - if the above conditions aren't met, pick a CPU that shares the same LLC
- *     to maintain cache locality.
+ *   - if the above conditions aren't met, pick a CPU that shares the same
+ *     LLC, if the LLC domain is a subset of @cpus_allowed, to maintain
+ *     cache locality.
  *
  * 4. Pick a CPU within the same NUMA node, if enabled:
- *   - choose a CPU from the same NUMA node to reduce memory access latency.
+ *   - choose a CPU from the same NUMA node, if the node cpumask is a
+ *     subset of @cpus_allowed, to reduce memory access latency.
  *
- * 5. Pick any idle CPU usable by the task.
+ * 5. Pick any idle CPU within the @cpus_allowed domain.
  *
  * Step 3 and 4 are performed only if the system has, respectively,
  * multiple LLCs / multiple NUMA nodes (see scx_selcpu_topo_llc and
@@ -464,9 +476,43 @@ s32 scx_select_cpu_dfl(struct task_struct *p, s32 prev_cpu, u64 wake_flags,
 		       const struct cpumask *cpus_allowed, u64 flags)
 {
 	const struct cpumask *llc_cpus = NULL, *numa_cpus = NULL;
-	int node = scx_cpu_node_if_enabled(prev_cpu);
+	const struct cpumask *allowed = p->cpus_ptr;
+	int node;
 	s32 cpu;
 
+	preempt_disable();
+
+	/*
+	 * Determine the subset of CPUs usable by @p within @cpus_allowed.
+	 */
+	if (cpus_allowed != p->cpus_ptr) {
+		struct cpumask *local_cpus = this_cpu_cpumask_var_ptr(local_idle_cpumask);
+
+		if (task_affinity_all(p) || cpumask_subset(cpus_allowed, p->cpus_ptr)) {
+			allowed = cpus_allowed;
+		} else if (cpumask_and(local_cpus, cpus_allowed, p->cpus_ptr)) {
+			allowed = local_cpus;
+		} else {
+			cpu = -EBUSY;
+			goto out_enable;
+		}
+	}
+
+	/*
+	 * If @prev_cpu is not in the allowed domain, try to assign a new
+	 * arbitrary CPU usable by the task in the allowed domain.
+	 */
+	if (!cpumask_test_cpu(prev_cpu, allowed)) {
+		cpu = cpumask_any_and_distribute(p->cpus_ptr, allowed);
+		if (cpu < nr_cpu_ids) {
+			prev_cpu = cpu;
+		} else {
+			cpu = -EBUSY;
+			goto out_enable;
+		}
+	}
+	node = scx_cpu_node_if_enabled(prev_cpu);
+
 	/*
 	 * This is necessary to protect llc_cpus.
 	 */
@@ -476,19 +522,13 @@ s32 scx_select_cpu_dfl(struct task_struct *p, s32 prev_cpu, u64 wake_flags,
 	 * Determine the subset of CPUs that the task can use in its
 	 * current LLC and node.
 	 */
-	if (static_branch_maybe(CONFIG_NUMA, &scx_selcpu_topo_numa)) {
-		numa_cpus = task_cpumask(p, cpus_allowed, numa_span(prev_cpu),
+	if (static_branch_maybe(CONFIG_NUMA, &scx_selcpu_topo_numa))
+		numa_cpus = task_cpumask(p, allowed, numa_span(prev_cpu),
 					 this_cpu_cpumask_var_ptr(local_numa_idle_cpumask));
-		if (cpumask_equal(numa_cpus, cpus_allowed))
-			numa_cpus = NULL;
-	}
 
-	if (static_branch_maybe(CONFIG_SCHED_MC, &scx_selcpu_topo_llc)) {
-		llc_cpus = task_cpumask(p, cpus_allowed, llc_span(prev_cpu),
+	if (static_branch_maybe(CONFIG_SCHED_MC, &scx_selcpu_topo_llc))
+		llc_cpus = task_cpumask(p, allowed, llc_span(prev_cpu),
 					this_cpu_cpumask_var_ptr(local_llc_idle_cpumask));
-		if (cpumask_equal(llc_cpus, cpus_allowed))
-			llc_cpus = NULL;
-	}
 
 	/*
 	 * If WAKE_SYNC, try to migrate the wakee to the waker's CPU.
@@ -525,7 +565,7 @@ s32 scx_select_cpu_dfl(struct task_struct *p, s32 prev_cpu, u64 wake_flags,
 		    cpu_rq(cpu)->scx.local_dsq.nr == 0 &&
 		    (!(flags & SCX_PICK_IDLE_IN_NODE) || (waker_node == node)) &&
 		    !cpumask_empty(idle_cpumask(waker_node)->cpu)) {
-			if (cpumask_test_cpu(cpu, cpus_allowed))
+			if (cpumask_test_cpu(cpu, allowed))
 				goto out_unlock;
 		}
 	}
@@ -570,7 +610,7 @@ s32 scx_select_cpu_dfl(struct task_struct *p, s32 prev_cpu, u64 wake_flags,
 		 * begin in prev_cpu's node and proceed to other nodes in
 		 * order of increasing distance.
 		 */
-		cpu = scx_pick_idle_cpu(cpus_allowed, node, flags | SCX_PICK_IDLE_CORE);
+		cpu = scx_pick_idle_cpu(allowed, node, flags | SCX_PICK_IDLE_CORE);
 		if (cpu >= 0)
 			goto out_unlock;
 
@@ -618,12 +658,14 @@ s32 scx_select_cpu_dfl(struct task_struct *p, s32 prev_cpu, u64 wake_flags,
 	 * in prev_cpu's node and proceed to other nodes in order of
 	 * increasing distance.
 	 */
-	cpu = scx_pick_idle_cpu(cpus_allowed, node, flags);
+	cpu = scx_pick_idle_cpu(allowed, node, flags);
 	if (cpu >= 0)
 		goto out_unlock;
 
 out_unlock:
 	rcu_read_unlock();
+out_enable:
+	preempt_enable();
 
 	return cpu;
 }
@@ -655,6 +697,8 @@ void scx_idle_init_masks(void)
 
 	/* Allocate local per-cpu idle cpumasks */
 	for_each_possible_cpu(i) {
+		BUG_ON(!alloc_cpumask_var_node(&per_cpu(local_idle_cpumask, i),
+					       GFP_KERNEL, cpu_to_node(i)));
 		BUG_ON(!alloc_cpumask_var_node(&per_cpu(local_llc_idle_cpumask, i),
 					       GFP_KERNEL, cpu_to_node(i)));
 		BUG_ON(!alloc_cpumask_var_node(&per_cpu(local_numa_idle_cpumask, i),
-- 
2.48.1