Merge tag 'sched_ext-for-6.15' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext

Pull sched_ext updates from Tejun Heo:

 - Add mechanism to count and report internal events. This significantly
   improves visibility on subtle corner conditions.

 - The default idle CPU selection logic is revamped and improved in
   multiple ways including being made topology aware.

 - sched_ext was disabling ttwu_queue for simplicity, which can be
   costly when hardware topology is more complex. Implement
   SCX_OPS_ALLOWED_QUEUED_WAKEUP so that BPF schedulers can selectively
   enable ttwu_queue.

 - tools/sched_ext updates to improve compatibility among others.

 - Other misc updates and fixes.

 - sched_ext/for-6.14-fixes were pulled a few times to receive
   prerequisite fixes and resolve conflicts.

* tag 'sched_ext-for-6.15' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext: (42 commits)
  sched_ext: idle: Refactor scx_select_cpu_dfl()
  sched_ext: idle: Honor idle flags in the built-in idle selection policy
  sched_ext: Skip per-CPU tasks in scx_bpf_reenqueue_local()
  sched_ext: Add trace point to track sched_ext core events
  sched_ext: Change the event type from u64 to s64
  sched_ext: Documentation: add task lifecycle summary
  tools/sched_ext: Provide a compatible helper for scx_bpf_events()
  selftests/sched_ext: Add NUMA-aware scheduler test
  tools/sched_ext: Provide consistent access to scx flags
  sched_ext: idle: Fix scx_bpf_pick_any_cpu_node() behavior
  sched_ext: idle: Introduce scx_bpf_nr_node_ids()
  sched_ext: idle: Introduce node-aware idle cpu kfunc helpers
  sched_ext: idle: Per-node idle cpumasks
  sched_ext: idle: Introduce SCX_OPS_BUILTIN_IDLE_PER_NODE
  sched_ext: idle: Make idle static keys private
  sched/topology: Introduce for_each_node_numadist() iterator
  mm/numa: Introduce nearest_node_nodemask()
  nodemask: numa: reorganize inclusion path
  nodemask: add nodes_copy()
  tools/sched_ext: Sync with scx repo
  ...

1 files changed, 95 insertions, 0 deletions

diff --git a/tools/sched_ext/include/scx/compat.bpf.h b/tools/sched_ext/include/scx/compat.bpf.h
index 50e1499ae093..9252e1a00556 100644
--- a/tools/sched_ext/include/scx/compat.bpf.h
+++ b/tools/sched_ext/include/scx/compat.bpf.h
@@ -125,12 +125,107 @@ bool scx_bpf_dispatch_vtime_from_dsq___compat(struct bpf_iter_scx_dsq *it__iter,
 	false;									\
 })
 
+/**
+ * __COMPAT_is_enq_cpu_selected - Test if SCX_ENQ_CPU_SELECTED is on
+ * in a compatible way. We will preserve this __COMPAT helper until v6.16.
+ *
+ * @enq_flags: enqueue flags from ops.enqueue()
+ *
+ * Return: True if SCX_ENQ_CPU_SELECTED is turned on in @enq_flags
+ */
+static inline bool __COMPAT_is_enq_cpu_selected(u64 enq_flags)
+{
+#ifdef HAVE_SCX_ENQ_CPU_SELECTED
+	/*
+	 * This is the case that a BPF code compiled against vmlinux.h
+	 * where the enum SCX_ENQ_CPU_SELECTED exists.
+	 */
+
+	/*
+	 * We should temporarily suspend the macro expansion of
+	 * 'SCX_ENQ_CPU_SELECTED'. This avoids 'SCX_ENQ_CPU_SELECTED' being
+	 * rewritten to '__SCX_ENQ_CPU_SELECTED' when 'SCX_ENQ_CPU_SELECTED'
+	 * is defined in 'scripts/gen_enums.py'.
+	 */
+#pragma push_macro("SCX_ENQ_CPU_SELECTED")
+#undef SCX_ENQ_CPU_SELECTED
+	u64 flag;
+
+	/*
+	 * When the kernel did not have SCX_ENQ_CPU_SELECTED,
+	 * select_task_rq_scx() has never been skipped. Thus, this case
+	 * should be considered that the CPU has already been selected.
+	 */
+	if (!bpf_core_enum_value_exists(enum scx_enq_flags,
+					SCX_ENQ_CPU_SELECTED))
+		return true;
+
+	flag = bpf_core_enum_value(enum scx_enq_flags, SCX_ENQ_CPU_SELECTED);
+	return enq_flags & flag;
+
+	/*
+	 * Once done, resume the macro expansion of 'SCX_ENQ_CPU_SELECTED'.
+	 */
+#pragma pop_macro("SCX_ENQ_CPU_SELECTED")
+#else
+	/*
+	 * This is the case that a BPF code compiled against vmlinux.h
+	 * where the enum SCX_ENQ_CPU_SELECTED does NOT exist.
+	 */
+	return true;
+#endif /* HAVE_SCX_ENQ_CPU_SELECTED */
+}
+
+
 #define scx_bpf_now()								\
 	(bpf_ksym_exists(scx_bpf_now) ?						\
 	 scx_bpf_now() :							\
 	 bpf_ktime_get_ns())
 
 /*
+ * v6.15: Introduce event counters.
+ *
+ * Preserve the following macro until v6.17.
+ */
+#define __COMPAT_scx_bpf_events(events, size)					\
+	(bpf_ksym_exists(scx_bpf_events) ?					\
+	 scx_bpf_events(events, size) : ({}))
+
+/*
+ * v6.15: Introduce NUMA-aware kfuncs to operate with per-node idle
+ * cpumasks.
+ *
+ * Preserve the following __COMPAT_scx_*_node macros until v6.17.
+ */
+#define __COMPAT_scx_bpf_nr_node_ids()						\
+	(bpf_ksym_exists(scx_bpf_nr_node_ids) ?					\
+	 scx_bpf_nr_node_ids() : 1U)
+
+#define __COMPAT_scx_bpf_cpu_node(cpu)						\
+	(bpf_ksym_exists(scx_bpf_cpu_node) ?					\
+	 scx_bpf_cpu_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(cpus_allowed, node, flags)		\
+	(bpf_ksym_exists(scx_bpf_pick_idle_cpu_node) ?				\
+	 scx_bpf_pick_idle_cpu_node(cpus_allowed, node, flags) :		\
+	 scx_bpf_pick_idle_cpu(cpus_allowed, flags))
+
+#define __COMPAT_scx_bpf_pick_any_cpu_node(cpus_allowed, node, flags)		\
+	(bpf_ksym_exists(scx_bpf_pick_any_cpu_node) ?				\
+	 scx_bpf_pick_any_cpu_node(cpus_allowed, node, flags) :			\
+	 scx_bpf_pick_any_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().
  */