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, 151 insertions, 0 deletions

diff --git a/net/bluetooth/hci_sysfs.c b/net/bluetooth/hci_sysfs.c
new file mode 100644
index 000000000..c2065abf7
--- /dev/null
+++ b/net/bluetooth/hci_sysfs.c
@@ -0,0 +1,151 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Bluetooth HCI driver model support. */
+
+#include <linux/module.h>
+
+#include <net/bluetooth/bluetooth.h>
+#include <net/bluetooth/hci_core.h>
+
+static const struct class bt_class = {
+	.name = "bluetooth",
+};
+
+static void bt_link_release(struct device *dev)
+{
+	struct hci_conn *conn = to_hci_conn(dev);
+	struct device *parent = dev->parent;
+
+	kfree(conn);
+	put_device(parent);
+}
+
+static const struct device_type bt_link = {
+	.name    = "link",
+	.release = bt_link_release,
+};
+
+/*
+ * The rfcomm tty device will possibly retain even when conn
+ * is down, and sysfs doesn't support move zombie device,
+ * so we should move the device before conn device is destroyed.
+ */
+static int __match_tty(struct device *dev, const void *data)
+{
+	return !strncmp(dev_name(dev), "rfcomm", 6);
+}
+
+void hci_conn_init_sysfs(struct hci_conn *conn)
+{
+	struct hci_dev *hdev = conn->hdev;
+
+	bt_dev_dbg(hdev, "conn %p", conn);
+
+	conn->dev.type = &bt_link;
+	conn->dev.class = &bt_class;
+	conn->dev.parent = get_device(&hdev->dev);
+
+	device_initialize(&conn->dev);
+}
+
+void hci_conn_add_sysfs(struct hci_conn *conn)
+{
+	struct hci_dev *hdev = conn->hdev;
+
+	bt_dev_dbg(hdev, "conn %p", conn);
+
+	if (device_is_registered(&conn->dev))
+		return;
+
+	dev_set_name(&conn->dev, "%s:%d", hdev->name, conn->handle);
+
+	if (device_add(&conn->dev) < 0)
+		bt_dev_err(hdev, "failed to register connection device");
+}
+
+void hci_conn_del_sysfs(struct hci_conn *conn)
+{
+	struct hci_dev *hdev = conn->hdev;
+
+	bt_dev_dbg(hdev, "conn %p", conn);
+
+	if (!device_is_registered(&conn->dev)) {
+		/* If device_add() has *not* succeeded, use *only* put_device()
+		 * to drop the reference count.
+		 */
+		put_device(&conn->dev);
+		return;
+	}
+
+	/* If there are devices using the connection as parent reset it to NULL
+	 * before unregistering the device.
+	 */
+	while (1) {
+		struct device *dev;
+
+		dev = device_find_child(&conn->dev, NULL, __match_tty);
+		if (!dev)
+			break;
+		device_move(dev, NULL, DPM_ORDER_DEV_LAST);
+		put_device(dev);
+	}
+
+	device_unregister(&conn->dev);
+}
+
+static void bt_host_release(struct device *dev)
+{
+	struct hci_dev *hdev = to_hci_dev(dev);
+
+	if (hci_dev_test_flag(hdev, HCI_UNREGISTER)) {
+		hci_release_dev(hdev);
+	} else {
+		cleanup_srcu_struct(&hdev->srcu);
+		kfree(hdev);
+	}
+	module_put(THIS_MODULE);
+}
+
+static ssize_t reset_store(struct device *dev, struct device_attribute *attr,
+			   const char *buf, size_t count)
+{
+	struct hci_dev *hdev = to_hci_dev(dev);
+
+	if (hdev->reset)
+		hdev->reset(hdev);
+
+	return count;
+}
+static DEVICE_ATTR_WO(reset);
+
+static struct attribute *bt_host_attrs[] = {
+	&dev_attr_reset.attr,
+	NULL,
+};
+ATTRIBUTE_GROUPS(bt_host);
+
+static const struct device_type bt_host = {
+	.name    = "host",
+	.release = bt_host_release,
+	.groups = bt_host_groups,
+};
+
+void hci_init_sysfs(struct hci_dev *hdev)
+{
+	struct device *dev = &hdev->dev;
+
+	dev->type = &bt_host;
+	dev->class = &bt_class;
+
+	__module_get(THIS_MODULE);
+	device_initialize(dev);
+}
+
+int __init bt_sysfs_init(void)
+{
+	return class_register(&bt_class);
+}
+
+void bt_sysfs_cleanup(void)
+{
+	class_unregister(&bt_class);
+}