sched/cache: Keep nr_pref_llc_running in the runnable domain, to fix LLC mis-scheduling buggrafted

alb_break_llc() decides whether to break LLC preference during active
load balance. It does so by testing that every runnable fair task on the
source rq prefers its LLC:

	env->src_rq->nr_pref_llc_running == env->src_rq->cfs.h_nr_runnable

But the two counters cover different sets. nr_pref_llc_running is updated
in account_llc_enqueue()/account_llc_dequeue(), next to cfs_rq->nr_queued,
so it follows queued tasks. h_nr_runnable is updated in set_delayed()/
clear_delayed() and drops delay-dequeued tasks.

So under DELAY_DEQUEUE, a preferring task that goes to sleep stays counted
in nr_pref_llc_running while h_nr_runnable falls. The equality then breaks,
alb_break_llc() returns false, and active balance is free to pull a task
off its preferred LLC. Active balance only moves runnable tasks, and this
is the only LLC check it consults: once the stopper runs, LBF_ACTIVE_LB
skips the per-task test in can_migrate_task(). The runnable set is the one
we want.

Fix it on the counter side. A task should be counted in
nr_pref_llc_running exactly while it is both queued on its preferred LLC
(pref_llc_queued) and runnable (!sched_delayed). Define that membership
once in task_pref_llc_runnable(), and adjust the counter only through
pref_llc_running_inc()/pref_llc_running_dec() from the four sites that
change either input: account_llc_enqueue(), account_llc_dequeue(),
set_delayed() and clear_delayed(). Gating every update on the same
predicate keeps the delay, wake and dequeue paths from double-counting
or underflowing; see the comments at those sites for the ordering.

nr_llc_running and sd->llc_counts are not touched and stay on queued
semantics.

Fixes: 714059f79ff0 ("sched/cache: Handle moving single tasks to/from their preferred LLC")
Closes: https://lore.kernel.org/lkml/20260827135000.735138-1-zhanxusheng@xiaomi.com/
Reported-by: Zhan Xusheng <zhanxusheng@xiaomi.com>
Suggested-by: Chen Yu <yu.c.chen@intel.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Kayra Cizmeci <kayracizmeci@gmail.com>
Cc: <stable@kernel.org> # v7.2.x
Link: https://patch.msgid.link/06af61afedac32e6477f57feb4d658f6c411c3af.1790035273.git.tim.c.chen@linux.intel.com

1 files changed, 143 insertions, 0 deletions

diff --git a/samples/ftrace/sample-trace-array.c b/samples/ftrace/sample-trace-array.c
new file mode 100644
index 000000000..414761610
--- /dev/null
+++ b/samples/ftrace/sample-trace-array.c
@@ -0,0 +1,143 @@
+// SPDX-License-Identifier: GPL-2.0-only
+#include <linux/module.h>
+#include <linux/kthread.h>
+#include <linux/trace.h>
+#include <linux/trace_events.h>
+#include <linux/timer.h>
+#include <linux/err.h>
+#include <linux/jiffies.h>
+#include <linux/workqueue.h>
+
+/*
+ * Any file that uses trace points, must include the header.
+ * But only one file, must include the header by defining
+ * CREATE_TRACE_POINTS first.  This will make the C code that
+ * creates the handles for the trace points.
+ */
+#define CREATE_TRACE_POINTS
+#include "sample-trace-array.h"
+
+struct trace_array *tr;
+static void mytimer_handler(struct timer_list *unused);
+static struct task_struct *simple_tsk;
+
+static void trace_work_fn(struct work_struct *work)
+{
+	/*
+	 * Disable tracing for event "sample_event".
+	 */
+	trace_array_set_clr_event(tr, "sample-subsystem", "sample_event",
+			false);
+}
+static DECLARE_WORK(trace_work, trace_work_fn);
+
+/*
+ * mytimer: Timer setup to disable tracing for event "sample_event". This
+ * timer is only for the purposes of the sample module to demonstrate access of
+ * Ftrace instances from within kernel.
+ */
+static DEFINE_TIMER(mytimer, mytimer_handler);
+
+static void mytimer_handler(struct timer_list *unused)
+{
+	schedule_work(&trace_work);
+}
+
+static void simple_thread_func(int count)
+{
+	set_current_state(TASK_INTERRUPTIBLE);
+	schedule_timeout(HZ);
+
+	/*
+	 * Printing count value using trace_array_printk() - trace_printk()
+	 * equivalent for the instance buffers.
+	 */
+	trace_array_printk(tr, _THIS_IP_, "trace_array_printk: count=%d\n",
+			count);
+	/*
+	 * Tracepoint for event "sample_event". This will print the
+	 * current value of count and current jiffies.
+	 */
+	trace_sample_event(count, jiffies);
+}
+
+static int simple_thread(void *arg)
+{
+	int count = 0;
+	unsigned long delay = msecs_to_jiffies(5000);
+
+	/*
+	 * Enable tracing for "sample_event".
+	 */
+	trace_array_set_clr_event(tr, "sample-subsystem", "sample_event", true);
+
+	/*
+	 * Adding timer - mytimer. This timer will disable tracing after
+	 * delay seconds.
+	 *
+	 */
+	add_timer(&mytimer);
+	mod_timer(&mytimer, jiffies+delay);
+
+	while (!kthread_should_stop())
+		simple_thread_func(count++);
+
+	timer_delete(&mytimer);
+	cancel_work_sync(&trace_work);
+
+	/*
+	 * trace_array_put() decrements the reference counter associated with
+	 * the trace array - "tr". We are done using the trace array, hence
+	 * decrement the reference counter so that it can be destroyed using
+	 * trace_array_destroy().
+	 */
+	trace_array_put(tr);
+
+	return 0;
+}
+
+static int __init sample_trace_array_init(void)
+{
+	/*
+	 * Return a pointer to the trace array with name "sample-instance" if it
+	 * exists, else create a new trace array.
+	 *
+	 * NOTE: This function increments the reference counter
+	 * associated with the trace array - "tr".
+	 */
+	tr = trace_array_get_by_name("sample-instance", "sched,timer,kprobes");
+
+	if (!tr)
+		return -1;
+	/*
+	 * If context specific per-cpu buffers havent already been allocated.
+	 */
+	trace_array_init_printk(tr);
+
+	simple_tsk = kthread_run(simple_thread, NULL, "sample-instance");
+	if (IS_ERR(simple_tsk)) {
+		trace_array_put(tr);
+		trace_array_destroy(tr);
+		return -1;
+	}
+
+	return 0;
+}
+
+static void __exit sample_trace_array_exit(void)
+{
+	kthread_stop(simple_tsk);
+
+	/*
+	 * We are unloading our module and no longer require the trace array.
+	 * Remove/destroy "tr" using trace_array_destroy()
+	 */
+	trace_array_destroy(tr);
+}
+
+module_init(sample_trace_array_init);
+module_exit(sample_trace_array_exit);
+
+MODULE_AUTHOR("Divya Indi");
+MODULE_DESCRIPTION("Sample module for kernel access to Ftrace instances");
+MODULE_LICENSE("GPL");