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, 80 insertions, 0 deletions

diff --git a/drivers/soc/apple/tunable.c b/drivers/soc/apple/tunable.c
new file mode 100644
index 000000000..659323839
--- /dev/null
+++ b/drivers/soc/apple/tunable.c
@@ -0,0 +1,80 @@
+// SPDX-License-Identifier: GPL-2.0-only OR MIT
+/*
+ * Apple Silicon hardware tunable support
+ *
+ * Each tunable is a list with each entry containing a offset into the MMIO
+ * region, a mask of bits to be cleared and a set of bits to be set. These
+ * tunables are passed along by the previous boot stages and vary from device
+ * to device such that they cannot be hardcoded in the individual drivers.
+ *
+ * Copyright (C) The Asahi Linux Contributors
+ */
+
+#include <linux/io.h>
+#include <linux/module.h>
+#include <linux/of.h>
+#include <linux/overflow.h>
+#include <linux/soc/apple/tunable.h>
+
+struct apple_tunable *devm_apple_tunable_parse(struct device *dev,
+					       struct device_node *np,
+					       const char *name,
+					       struct resource *res)
+{
+	struct apple_tunable *tunable;
+	struct property *prop;
+	const __be32 *p;
+	size_t sz;
+	int i;
+
+	if (resource_size(res) < 4)
+		return ERR_PTR(-EINVAL);
+
+	prop = of_find_property(np, name, NULL);
+	if (!prop)
+		return ERR_PTR(-ENOENT);
+
+	if (prop->length % (3 * sizeof(u32)))
+		return ERR_PTR(-EINVAL);
+	sz = prop->length / (3 * sizeof(u32));
+
+	tunable = devm_kzalloc(dev, struct_size(tunable, values, sz), GFP_KERNEL);
+	if (!tunable)
+		return ERR_PTR(-ENOMEM);
+	tunable->sz = sz;
+
+	for (i = 0, p = NULL; i < tunable->sz; ++i) {
+		p = of_prop_next_u32(prop, p, &tunable->values[i].offset);
+		p = of_prop_next_u32(prop, p, &tunable->values[i].mask);
+		p = of_prop_next_u32(prop, p, &tunable->values[i].value);
+
+		/* Sanity checks to catch bugs in our bootloader */
+		if (tunable->values[i].offset % 4)
+			return ERR_PTR(-EINVAL);
+		if (tunable->values[i].offset > (resource_size(res) - 4))
+			return ERR_PTR(-EINVAL);
+	}
+
+	return tunable;
+}
+EXPORT_SYMBOL(devm_apple_tunable_parse);
+
+void apple_tunable_apply(void __iomem *regs, struct apple_tunable *tunable)
+{
+	size_t i;
+
+	for (i = 0; i < tunable->sz; ++i) {
+		u32 val, old_val;
+
+		old_val = readl(regs + tunable->values[i].offset);
+		val = old_val & ~tunable->values[i].mask;
+		val |= tunable->values[i].value;
+		if (val != old_val)
+			writel(val, regs + tunable->values[i].offset);
+	}
+}
+EXPORT_SYMBOL(apple_tunable_apply);
+
+MODULE_LICENSE("Dual MIT/GPL");
+MODULE_AUTHOR("Sven Peter <sven@kernel.org>");
+MODULE_DESCRIPTION("Apple Silicon hardware tunable support");