Merge tag 'sched_ext-for-6.12-rc1-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext

Pull sched_ext fixes from Tejun Heo:

 - Three build fixes

 - The fix for a stall bug introduced by a recent optimization in sched
   core (SM_IDLE)

 - Addition of /sys/kernel/sched_ext/enable_seq. While not a fix, it is
   a simple addition that distro people want to be able to tell whether
   an SCX scheduler has ever been loaded on the system

* tag 'sched_ext-for-6.12-rc1-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext:
  sched_ext: Provide a sysfs enable_seq counter
  sched_ext: Fix build when !CONFIG_STACKTRACE
  sched, sched_ext: Disable SM_IDLE/rq empty path when scx_enabled()
  sched: Put task_group::idle under CONFIG_GROUP_SCHED_WEIGHT
  sched: Add dummy version of sched_group_set_idle()

Author: torvalds

Documentation/scheduler/sched-ext.rst

@@ -83,6 +83,15 @@ The current status of the BPF scheduler can be determined as follows:
     # cat /sys/kernel/sched_ext/root/ops
     simple
 
+You can check if any BPF scheduler has ever been loaded since boot by examining
+this monotonically incrementing counter (a value of zero indicates that no BPF
+scheduler has been loaded):
+
+.. code-block:: none
+
+    # cat /sys/kernel/sched_ext/enable_seq
+    1
+
 ``tools/sched_ext/scx_show_state.py`` is a drgn script which shows more
 detailed information:
 
@@ -96,6 +105,7 @@ detailed information:
     enable_state  : enabled (2)
     bypass_depth  : 0
     nr_rejected   : 0
+    enable_seq    : 1
 
 If ``CONFIG_SCHED_DEBUG`` is set, whether a given task is on sched_ext can
 be determined as follows:

kernel/sched/core.c

@@ -6591,7 +6591,8 @@ static void __sched notrace __schedule(int sched_mode)
 	 */
 	prev_state = READ_ONCE(prev->__state);
 	if (sched_mode == SM_IDLE) {
-		if (!rq->nr_running) {
+		/* SCX must consult the BPF scheduler to tell if rq is empty */
+		if (!rq->nr_running && !scx_enabled()) {
 			next = prev;
 			goto picked;
 		}

kernel/sched/ext.c

@@ -874,6 +874,13 @@ static struct scx_exit_info *scx_exit_info;
 static atomic_long_t scx_nr_rejected = ATOMIC_LONG_INIT(0);
 static atomic_long_t scx_hotplug_seq = ATOMIC_LONG_INIT(0);
 
+/*
+ * A monotically increasing sequence number that is incremented every time a
+ * scheduler is enabled. This can be used by to check if any custom sched_ext
+ * scheduler has ever been used in the system.
+ */
+static atomic_long_t scx_enable_seq = ATOMIC_LONG_INIT(0);
+
 /*
  * The maximum amount of time in jiffies that a task may be runnable without
  * being scheduled on a CPU. If this timeout is exceeded, it will trigger
@@ -4154,11 +4161,19 @@ static ssize_t scx_attr_hotplug_seq_show(struct kobject *kobj,
 }
 SCX_ATTR(hotplug_seq);
 
+static ssize_t scx_attr_enable_seq_show(struct kobject *kobj,
+					struct kobj_attribute *ka, char *buf)
+{
+	return sysfs_emit(buf, "%ld\n", atomic_long_read(&scx_enable_seq));
+}
+SCX_ATTR(enable_seq);
+
 static struct attribute *scx_global_attrs[] = {
 	&scx_attr_state.attr,
 	&scx_attr_switch_all.attr,
 	&scx_attr_nr_rejected.attr,
 	&scx_attr_hotplug_seq.attr,
+	&scx_attr_enable_seq.attr,
 	NULL,
 };
 
@@ -4469,8 +4484,9 @@ static void scx_ops_disable_workfn(struct kthread_work *work)
 
 		if (ei->msg[0] != '\0')
 			pr_err("sched_ext: %s: %s\n", scx_ops.name, ei->msg);
-
+#ifdef CONFIG_STACKTRACE
 		stack_trace_print(ei->bt, ei->bt_len, 2);
+#endif
 	} else {
 		pr_info("sched_ext: BPF scheduler \"%s\" disabled (%s)\n",
 			scx_ops.name, ei->reason);
@@ -4847,10 +4863,10 @@ static __printf(3, 4) void scx_ops_exit_kind(enum scx_exit_kind kind,
 		return;
 
 	ei->exit_code = exit_code;
-
+#ifdef CONFIG_STACKTRACE
 	if (kind >= SCX_EXIT_ERROR)
 		ei->bt_len = stack_trace_save(ei->bt, SCX_EXIT_BT_LEN, 1);
-
+#endif
 	va_start(args, fmt);
 	vscnprintf(ei->msg, SCX_EXIT_MSG_LEN, fmt, args);
 	va_end(args);
@@ -5176,6 +5192,8 @@ static int scx_ops_enable(struct sched_ext_ops *ops, struct bpf_link *link)
 	kobject_uevent(scx_root_kobj, KOBJ_ADD);
 	mutex_unlock(&scx_ops_enable_mutex);
 
+	atomic_long_inc(&scx_enable_seq);
+
 	return 0;
 
 err_del:

kernel/sched/sched.h

@@ -432,16 +432,17 @@ struct cfs_bandwidth {
 struct task_group {
 	struct cgroup_subsys_state css;
 
+#ifdef CONFIG_GROUP_SCHED_WEIGHT
+	/* A positive value indicates that this is a SCHED_IDLE group. */
+	int			idle;
+#endif
+
 #ifdef CONFIG_FAIR_GROUP_SCHED
 	/* schedulable entities of this group on each CPU */
 	struct sched_entity	**se;
 	/* runqueue "owned" by this group on each CPU */
 	struct cfs_rq		**cfs_rq;
 	unsigned long		shares;
-
-	/* A positive value indicates that this is a SCHED_IDLE group. */
-	int			idle;
-
 #ifdef	CONFIG_SMP
 	/*
 	 * load_avg can be heavily contended at clock tick time, so put
@@ -582,6 +583,7 @@ static inline void set_task_rq_fair(struct sched_entity *se,
 #endif /* CONFIG_SMP */
 #else /* !CONFIG_FAIR_GROUP_SCHED */
 static inline int sched_group_set_shares(struct task_group *tg, unsigned long shares) { return 0; }
+static inline int sched_group_set_idle(struct task_group *tg, long idle) { return 0; }
 #endif /* CONFIG_FAIR_GROUP_SCHED */
 
 #else /* CONFIG_CGROUP_SCHED */

tools/sched_ext/scx_show_state.py

@@ -37,3 +37,4 @@ def ops_state_str(state):
 print(f'enable_state  : {ops_state_str(enable_state)} ({enable_state})')
 print(f'bypass_depth  : {read_atomic("scx_ops_bypass_depth")}')
 print(f'nr_rejected   : {read_atomic("scx_nr_rejected")}')
+print(f'enable_seq    : {read_atomic("scx_enable_seq")}')