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, 127 insertions, 0 deletions

diff --git a/include/soc/tegra/fuse.h b/include/soc/tegra/fuse.h
new file mode 100644
index 000000000..c4f7a1b97
--- /dev/null
+++ b/include/soc/tegra/fuse.h
@@ -0,0 +1,127 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Copyright (c) 2012-2023, NVIDIA CORPORATION.  All rights reserved.
+ */
+
+#ifndef __SOC_TEGRA_FUSE_H__
+#define __SOC_TEGRA_FUSE_H__
+
+#include <linux/types.h>
+
+#define TEGRA20		0x20
+#define TEGRA30		0x30
+#define TEGRA114	0x35
+#define TEGRA124	0x40
+#define TEGRA132	0x13
+#define TEGRA210	0x21
+#define TEGRA186	0x18
+#define TEGRA194	0x19
+#define TEGRA234	0x23
+#define TEGRA241	0x24
+#define TEGRA264	0x26
+
+#define TEGRA_FUSE_SKU_CALIB_0	0xf0
+#define TEGRA30_FUSE_SATA_CALIB	0x124
+#define TEGRA_FUSE_USB_CALIB_EXT_0 0x250
+
+#ifndef __ASSEMBLER__
+
+enum tegra_revision {
+	TEGRA_REVISION_UNKNOWN = 0,
+	TEGRA_REVISION_A01,
+	TEGRA_REVISION_A02,
+	TEGRA_REVISION_A03,
+	TEGRA_REVISION_A03p,
+	TEGRA_REVISION_A04,
+	TEGRA_REVISION_MAX,
+};
+
+enum tegra_platform {
+	TEGRA_PLATFORM_SILICON = 0,
+	TEGRA_PLATFORM_QT,
+	TEGRA_PLATFORM_SYSTEM_FPGA,
+	TEGRA_PLATFORM_UNIT_FPGA,
+	TEGRA_PLATFORM_ASIM_QT,
+	TEGRA_PLATFORM_ASIM_LINSIM,
+	TEGRA_PLATFORM_DSIM_ASIM_LINSIM,
+	TEGRA_PLATFORM_VERIFICATION_SIMULATION,
+	TEGRA_PLATFORM_VDK,
+	TEGRA_PLATFORM_VSP,
+	TEGRA_PLATFORM_MAX,
+};
+
+struct tegra_sku_info {
+	int sku_id;
+	int cpu_process_id;
+	int cpu_speedo_id;
+	int cpu_speedo_value;
+	int cpu_iddq_value;
+	int soc_process_id;
+	int soc_speedo_id;
+	int soc_speedo_value;
+	int gpu_process_id;
+	int gpu_speedo_id;
+	int gpu_speedo_value;
+	enum tegra_revision revision;
+	enum tegra_platform platform;
+};
+
+#ifdef CONFIG_ARCH_TEGRA
+extern struct tegra_sku_info tegra_sku_info;
+u32 tegra_read_straps(void);
+u32 tegra_read_ram_code(void);
+int tegra_fuse_readl(unsigned long offset, u32 *value);
+u32 tegra_read_chipid(void);
+u8 tegra_get_chip_id(void);
+u8 tegra_get_platform(void);
+bool tegra_is_silicon(void);
+int tegra194_miscreg_mask_serror(void);
+#else
+static struct tegra_sku_info tegra_sku_info __maybe_unused;
+
+static inline u32 tegra_read_straps(void)
+{
+	return 0;
+}
+
+static inline u32 tegra_read_ram_code(void)
+{
+	return 0;
+}
+
+static inline int tegra_fuse_readl(unsigned long offset, u32 *value)
+{
+	return -ENODEV;
+}
+
+static inline u32 tegra_read_chipid(void)
+{
+	return 0;
+}
+
+static inline u8 tegra_get_chip_id(void)
+{
+	return 0;
+}
+
+static inline u8 tegra_get_platform(void)
+{
+	return 0;
+}
+
+static inline bool tegra_is_silicon(void)
+{
+	return false;
+}
+
+static inline int tegra194_miscreg_mask_serror(void)
+{
+	return false;
+}
+#endif
+
+struct device *tegra_soc_device_register(void);
+
+#endif /* __ASSEMBLER__ */
+
+#endif /* __SOC_TEGRA_FUSE_H__ */