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, 36 insertions, 0 deletions

diff --git a/Documentation/scheduler/sched-ext.rst b/Documentation/scheduler/sched-ext.rst
index c4672d7df2f7..0993e41353db 100644
--- a/Documentation/scheduler/sched-ext.rst
+++ b/Documentation/scheduler/sched-ext.rst
@@ -294,6 +294,42 @@ dispatching, and must be dispatched to with ``scx_bpf_dsq_insert()``. See
 the function documentation and usage in ``tools/sched_ext/scx_simple.bpf.c``
 for more information.
 
+Task Lifecycle
+--------------
+
+The following pseudo-code summarizes the entire lifecycle of a task managed
+by a sched_ext scheduler:
+
+.. code-block:: c
+
+    ops.init_task();            /* A new task is created */
+    ops.enable();               /* Enable BPF scheduling for the task */
+
+    while (task in SCHED_EXT) {
+        if (task can migrate)
+            ops.select_cpu();   /* Called on wakeup (optimization) */
+
+        ops.runnable();         /* Task becomes ready to run */
+
+        while (task is runnable) {
+            if (task is not in a DSQ) {
+                ops.enqueue();  /* Task can be added to a DSQ */
+
+                /* A CPU becomes available */
+
+                ops.dispatch(); /* Task is moved to a local DSQ */
+            }
+            ops.running();      /* Task starts running on its assigned CPU */
+            ops.tick();         /* Called every 1/HZ seconds */
+            ops.stopping();     /* Task stops running (time slice expires or wait) */
+        }
+
+        ops.quiescent();        /* Task releases its assigned CPU (wait) */
+    }
+
+    ops.disable();              /* Disable BPF scheduling for the task */
+    ops.exit_task();            /* Task is destroyed */
+
 Where to Look
 =============