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, 49 insertions, 0 deletions

diff --git a/drivers/mfd/ocelot.h b/drivers/mfd/ocelot.h
new file mode 100644
index 000000000..b8bc2f148
--- /dev/null
+++ b/drivers/mfd/ocelot.h
@@ -0,0 +1,49 @@
+/* SPDX-License-Identifier: GPL-2.0 OR MIT */
+/* Copyright 2021, 2022 Innovative Advantage Inc. */
+
+#ifndef _MFD_OCELOT_H
+#define _MFD_OCELOT_H
+
+#include <linux/kconfig.h>
+
+struct device;
+struct regmap;
+struct resource;
+
+/**
+ * struct ocelot_ddata - Private data for an external Ocelot chip
+ * @gcb_regmap:		General Configuration Block regmap. Used for
+ *			operations like chip reset.
+ * @cpuorg_regmap:	CPU Device Origin Block regmap. Used for operations
+ *			like SPI bus configuration.
+ * @spi_padding_bytes:	Number of padding bytes that must be thrown out before
+ *			read data gets returned. This is calculated during
+ *			initialization based on bus speed.
+ * @dummy_buf:		Zero-filled buffer of spi_padding_bytes size. The dummy
+ *			bytes that will be sent out between the address and
+ *			data of a SPI read operation.
+ */
+struct ocelot_ddata {
+	struct regmap *gcb_regmap;
+	struct regmap *cpuorg_regmap;
+	int spi_padding_bytes;
+	void *dummy_buf;
+};
+
+int ocelot_chip_reset(struct device *dev);
+int ocelot_core_init(struct device *dev);
+
+/* SPI-specific routines that won't be necessary for other interfaces */
+struct regmap *ocelot_spi_init_regmap(struct device *dev,
+				      const struct resource *res);
+
+#define OCELOT_SPI_BYTE_ORDER_LE 0x00000000
+#define OCELOT_SPI_BYTE_ORDER_BE 0x81818181
+
+#ifdef __LITTLE_ENDIAN
+#define OCELOT_SPI_BYTE_ORDER OCELOT_SPI_BYTE_ORDER_LE
+#else
+#define OCELOT_SPI_BYTE_ORDER OCELOT_SPI_BYTE_ORDER_BE
+#endif
+
+#endif