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/drivers/acpi/proc.c b/drivers/acpi/proc.c
new file mode 100644
index 000000000..c08ead072
--- /dev/null
+++ b/drivers/acpi/proc.c
@@ -0,0 +1,143 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <linux/proc_fs.h>
+#include <linux/seq_file.h>
+#include <linux/string_choices.h>
+#include <linux/suspend.h>
+#include <linux/bcd.h>
+#include <linux/acpi.h>
+#include <linux/uaccess.h>
+
+#include "sleep.h"
+#include "internal.h"
+
+/*
+ * this file provides support for:
+ * /proc/acpi/wakeup
+ */
+
+static int
+acpi_system_wakeup_device_seq_show(struct seq_file *seq, void *offset)
+{
+	struct acpi_device *dev, *tmp;
+
+	seq_printf(seq, "Device\tS-state\t  Status   Sysfs node\n");
+
+	mutex_lock(&acpi_device_lock);
+	list_for_each_entry_safe(dev, tmp, &acpi_wakeup_device_list,
+				 wakeup_list) {
+		struct acpi_device_physical_node *entry;
+
+		if (!dev->wakeup.flags.valid)
+			continue;
+
+		seq_printf(seq, "%s\t  S%llu\t",
+			   dev->pnp.bus_id,
+			   dev->wakeup.sleep_state);
+
+		mutex_lock(&dev->physical_node_lock);
+
+		if (!dev->physical_node_count) {
+			seq_printf(seq, "%c%-8s\n",
+				dev->wakeup.flags.valid ? '*' : ' ',
+				str_enabled_disabled(device_may_wakeup(&dev->dev)));
+		} else {
+			struct device *ldev;
+			list_for_each_entry(entry, &dev->physical_node_list,
+					node) {
+				ldev = get_device(entry->dev);
+				if (!ldev)
+					continue;
+
+				if (&entry->node !=
+						dev->physical_node_list.next)
+					seq_printf(seq, "\t\t");
+
+				seq_printf(seq, "%c%-8s  %s:%s\n",
+					dev->wakeup.flags.valid ? '*' : ' ',
+					str_enabled_disabled(device_may_wakeup(ldev) ||
+							     device_may_wakeup(&dev->dev)),
+					ldev->bus ? ldev->bus->name :
+					"no-bus", dev_name(ldev));
+				put_device(ldev);
+			}
+		}
+
+		mutex_unlock(&dev->physical_node_lock);
+	}
+	mutex_unlock(&acpi_device_lock);
+	return 0;
+}
+
+static void physical_device_enable_wakeup(struct acpi_device *adev)
+{
+	struct acpi_device_physical_node *entry;
+
+	mutex_lock(&adev->physical_node_lock);
+
+	list_for_each_entry(entry,
+		&adev->physical_node_list, node)
+		if (entry->dev && device_can_wakeup(entry->dev)) {
+			bool enable = !device_may_wakeup(entry->dev);
+			device_set_wakeup_enable(entry->dev, enable);
+		}
+
+	mutex_unlock(&adev->physical_node_lock);
+}
+
+static ssize_t
+acpi_system_write_wakeup_device(struct file *file,
+				const char __user * buffer,
+				size_t count, loff_t * ppos)
+{
+	struct acpi_device *dev, *tmp;
+	char strbuf[5];
+	char str[5] = "";
+
+	if (count > 4)
+		count = 4;
+
+	if (copy_from_user(strbuf, buffer, count))
+		return -EFAULT;
+	strbuf[count] = '\0';
+	sscanf(strbuf, "%s", str);
+
+	mutex_lock(&acpi_device_lock);
+	list_for_each_entry_safe(dev, tmp, &acpi_wakeup_device_list,
+				 wakeup_list) {
+		if (!dev->wakeup.flags.valid)
+			continue;
+
+		if (!strncmp(dev->pnp.bus_id, str, 4)) {
+			if (device_can_wakeup(&dev->dev)) {
+				bool enable = !device_may_wakeup(&dev->dev);
+				device_set_wakeup_enable(&dev->dev, enable);
+			} else {
+				physical_device_enable_wakeup(dev);
+			}
+			break;
+		}
+	}
+	mutex_unlock(&acpi_device_lock);
+	return count;
+}
+
+static int
+acpi_system_wakeup_device_open_fs(struct inode *inode, struct file *file)
+{
+	return single_open(file, acpi_system_wakeup_device_seq_show,
+			   pde_data(inode));
+}
+
+static const struct proc_ops acpi_system_wakeup_device_proc_ops = {
+	.proc_open	= acpi_system_wakeup_device_open_fs,
+	.proc_read	= seq_read,
+	.proc_write	= acpi_system_write_wakeup_device,
+	.proc_lseek	= seq_lseek,
+	.proc_release	= single_release,
+};
+
+void __init acpi_sleep_proc_init(void)
+{
+	/* 'wakeup device' [R/W] */
+	proc_create("wakeup", 0644, acpi_root_dir, &acpi_system_wakeup_device_proc_ops);
+}