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, 318 insertions, 0 deletions

diff --git a/include/acpi/cppc_acpi.h b/include/acpi/cppc_acpi.h
new file mode 100644
index 000000000..94a6277ed
--- /dev/null
+++ b/include/acpi/cppc_acpi.h
@@ -0,0 +1,318 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * CPPC (Collaborative Processor Performance Control) methods used
+ * by CPUfreq drivers.
+ *
+ * (C) Copyright 2014, 2015 Linaro Ltd.
+ * Author: Ashwin Chaugule <ashwin.chaugule@linaro.org>
+ */
+
+#ifndef _CPPC_ACPI_H
+#define _CPPC_ACPI_H
+
+#include <linux/acpi.h>
+#include <linux/cpufreq.h>
+#include <linux/types.h>
+
+#include <acpi/pcc.h>
+#include <acpi/processor.h>
+
+/* CPPCv2, CPPCv3 and CPPCv4 support */
+#define CPPC_V2_REV	2
+#define CPPC_V3_REV	3
+#define CPPC_V4_REV	4
+#define CPPC_V2_NUM_ENT	21
+#define CPPC_V3_NUM_ENT	23
+#define CPPC_V4_NUM_ENT	25
+
+#define PCC_CMD_COMPLETE_MASK	(1 << 0)
+#define PCC_ERROR_MASK		(1 << 2)
+
+#define MAX_CPC_REG_ENT 23
+
+/* CPPC specific PCC commands. */
+#define	CMD_READ 0
+#define	CMD_WRITE 1
+
+#define CPPC_AUTO_ACT_WINDOW_SIG_BIT_SIZE	(7)
+#define CPPC_AUTO_ACT_WINDOW_EXP_BIT_SIZE	(3)
+#define CPPC_AUTO_ACT_WINDOW_MAX_SIG	((1 << CPPC_AUTO_ACT_WINDOW_SIG_BIT_SIZE) - 1)
+#define CPPC_AUTO_ACT_WINDOW_MAX_EXP	((1 << CPPC_AUTO_ACT_WINDOW_EXP_BIT_SIZE) - 1)
+/* CPPC_AUTO_ACT_WINDOW_MAX_SIG is 127, so 128 and 129 will decay to 127 when writing */
+#define CPPC_AUTO_ACT_WINDOW_SIG_CARRY_THRESH 129
+
+#define CPPC_EPP_PERFORMANCE_PREF		0x00
+#define CPPC_EPP_ENERGY_EFFICIENCY_PREF		0xFF
+
+#define CPPC_PERF_LIMITED_DESIRED_EXCURSION	BIT(0)
+#define CPPC_PERF_LIMITED_MINIMUM_EXCURSION	BIT(1)
+#define CPPC_PERF_LIMITED_MASK		(CPPC_PERF_LIMITED_DESIRED_EXCURSION | \
+					 CPPC_PERF_LIMITED_MINIMUM_EXCURSION)
+
+/* Each register has the folowing format. */
+struct cpc_reg {
+	u8 descriptor;
+	u16 length;
+	u8 space_id;
+	u8 bit_width;
+	u8 bit_offset;
+	u8 access_width;
+	u64 address;
+} __packed;
+
+/*
+ * Each entry in the CPC table is either
+ * of type ACPI_TYPE_BUFFER or
+ * ACPI_TYPE_INTEGER.
+ */
+struct cpc_register_resource {
+	acpi_object_type type;
+	u64 __iomem *sys_mem_vaddr;
+	union {
+		struct {
+			struct cpc_reg reg;
+			bool use_rmw_lock;
+		};
+		u64 int_value;
+	} cpc_entry;
+};
+
+/* Container to hold the CPC details for each CPU */
+struct cpc_desc {
+	int num_entries;
+	int version;
+	int cpu_id;
+	int write_cmd_status;
+	int write_cmd_id;
+	/* Lock used for RMW operations in cpc_write() */
+	raw_spinlock_t rmw_lock;
+	struct cpc_register_resource cpc_regs[MAX_CPC_REG_ENT];
+	struct acpi_psd_package domain_info;
+	struct kobject kobj;
+};
+
+/* These are indexes into the per-cpu cpc_regs[]. Order is important. */
+enum cppc_regs {
+	HIGHEST_PERF,
+	NOMINAL_PERF,
+	LOW_NON_LINEAR_PERF,
+	LOWEST_PERF,
+	GUARANTEED_PERF,
+	DESIRED_PERF,
+	MIN_PERF,
+	MAX_PERF,
+	PERF_REDUC_TOLERANCE,
+	TIME_WINDOW,
+	CTR_WRAP_TIME,
+	REFERENCE_CTR,
+	DELIVERED_CTR,
+	PERF_LIMITED,
+	ENABLE,
+	AUTO_SEL_ENABLE,
+	AUTO_ACT_WINDOW,
+	ENERGY_PERF,
+	REFERENCE_PERF,
+	LOWEST_FREQ,
+	NOMINAL_FREQ,
+	OSPM_NOMINAL_PERF,
+	RESOURCE_PRIORITY,
+};
+
+/*
+ * Categorization of registers as described
+ * in the ACPI v.5.1 spec.
+ * XXX: Only filling up ones which are used by governors
+ * today.
+ */
+struct cppc_perf_caps {
+	u32 guaranteed_perf;
+	u32 highest_perf;
+	u32 nominal_perf;
+	u32 reference_perf;
+	u32 lowest_perf;
+	u32 lowest_nonlinear_perf;
+	u32 lowest_freq;
+	u32 nominal_freq;
+};
+
+struct cppc_perf_ctrls {
+	u32 max_perf;
+	u32 min_perf;
+	u32 desired_perf;
+	u32 energy_perf;
+	bool auto_sel;
+};
+
+struct cppc_perf_fb_ctrs {
+	u64 reference;
+	u64 delivered;
+	u64 wraparound_time;
+};
+
+/* Per CPU container for runtime CPPC management. */
+struct cppc_cpudata {
+	struct cppc_perf_caps perf_caps;
+	struct cppc_perf_ctrls perf_ctrls;
+	struct cppc_perf_fb_ctrs perf_fb_ctrs;
+	unsigned int shared_type;
+	cpumask_var_t shared_cpu_map;
+};
+
+#ifdef CONFIG_ACPI_CPPC_LIB
+extern int cppc_get_desired_perf(int cpunum, u64 *desired_perf);
+extern int cppc_get_nominal_perf(int cpunum, u64 *nominal_perf);
+extern int cppc_get_highest_perf(int cpunum, u64 *highest_perf);
+extern int cppc_get_perf_ctrs(int cpu, struct cppc_perf_fb_ctrs *perf_fb_ctrs);
+extern int cppc_get_perf(int cpu, struct cppc_perf_ctrls *perf_ctrls);
+extern int cppc_set_perf(int cpu, struct cppc_perf_ctrls *perf_ctrls);
+extern int cppc_set_enable(int cpu, bool enable);
+extern int cppc_get_perf_caps(int cpu, struct cppc_perf_caps *caps);
+extern bool cppc_perf_ctrs_in_pcc_cpu(unsigned int cpu);
+extern bool cppc_perf_ctrs_in_pcc(void);
+extern u64 cppc_get_dmi_max_khz(void);
+extern unsigned int cppc_perf_to_khz(struct cppc_perf_caps *caps, unsigned int perf);
+extern unsigned int cppc_khz_to_perf(struct cppc_perf_caps *caps, unsigned int freq);
+extern bool acpi_cpc_valid(void);
+bool cppc_allow_fast_switch(const struct cpumask *cpus);
+extern int acpi_get_psd_map(unsigned int cpu, struct cppc_cpudata *cpu_data);
+extern int cppc_get_transition_latency(int cpu);
+extern bool cpc_ffh_supported(void);
+extern bool cpc_supported_by_cpu(void);
+extern int cpc_read_ffh(int cpunum, struct cpc_reg *reg, u64 *val);
+extern int cpc_read_ffh_fb_ctrs(int cpu, struct cpc_reg *reg1, u64 *val1,
+				struct cpc_reg *reg2, u64 *val2);
+extern int cpc_write_ffh(int cpunum, struct cpc_reg *reg, u64 val);
+extern int cppc_get_epp_perf(int cpunum, u64 *epp_perf);
+extern int cppc_set_epp_perf(int cpu, struct cppc_perf_ctrls *perf_ctrls, bool enable);
+extern int cppc_set_epp(int cpu, u64 epp_val);
+extern int cppc_get_auto_act_window(int cpu, u64 *auto_act_window);
+extern int cppc_set_auto_act_window(int cpu, u64 auto_act_window);
+extern int cppc_get_auto_sel(int cpu, bool *enable);
+extern int cppc_set_auto_sel(int cpu, bool enable);
+extern int cppc_get_perf_limited(int cpu, u64 *perf_limited);
+extern int cppc_set_perf_limited(int cpu, u64 bits_to_clear);
+extern int amd_get_highest_perf(unsigned int cpu, u32 *highest_perf);
+extern int amd_get_boost_ratio_numerator(unsigned int cpu, u64 *numerator);
+extern int amd_detect_prefcore(bool *detected);
+#else /* !CONFIG_ACPI_CPPC_LIB */
+static inline int cppc_get_desired_perf(int cpunum, u64 *desired_perf)
+{
+	return -EOPNOTSUPP;
+}
+static inline int cppc_get_nominal_perf(int cpunum, u64 *nominal_perf)
+{
+	return -EOPNOTSUPP;
+}
+static inline int cppc_get_highest_perf(int cpunum, u64 *highest_perf)
+{
+	return -EOPNOTSUPP;
+}
+static inline int cppc_get_perf_ctrs(int cpu, struct cppc_perf_fb_ctrs *perf_fb_ctrs)
+{
+	return -EOPNOTSUPP;
+}
+static inline int cppc_get_perf(int cpu, struct cppc_perf_ctrls *perf_ctrls)
+{
+	return -EOPNOTSUPP;
+}
+static inline int cppc_set_perf(int cpu, struct cppc_perf_ctrls *perf_ctrls)
+{
+	return -EOPNOTSUPP;
+}
+static inline int cppc_set_enable(int cpu, bool enable)
+{
+	return -EOPNOTSUPP;
+}
+static inline int cppc_get_perf_caps(int cpu, struct cppc_perf_caps *caps)
+{
+	return -EOPNOTSUPP;
+}
+static inline bool cppc_perf_ctrs_in_pcc_cpu(unsigned int cpu)
+{
+	return false;
+}
+static inline bool cppc_perf_ctrs_in_pcc(void)
+{
+	return false;
+}
+static inline bool acpi_cpc_valid(void)
+{
+	return false;
+}
+
+static inline bool cppc_allow_fast_switch(const struct cpumask *cpus)
+{
+	return false;
+}
+static inline int cppc_get_transition_latency(int cpu)
+{
+	return -ENODATA;
+}
+static inline bool cpc_ffh_supported(void)
+{
+	return false;
+}
+static inline int cpc_read_ffh(int cpunum, struct cpc_reg *reg, u64 *val)
+{
+	return -EOPNOTSUPP;
+}
+static inline int cpc_read_ffh_fb_ctrs(int cpu, struct cpc_reg *reg1, u64 *val1,
+				       struct cpc_reg *reg2, u64 *val2)
+{
+	return -EOPNOTSUPP;
+}
+static inline int cpc_write_ffh(int cpunum, struct cpc_reg *reg, u64 val)
+{
+	return -EOPNOTSUPP;
+}
+static inline int cppc_set_epp_perf(int cpu, struct cppc_perf_ctrls *perf_ctrls, bool enable)
+{
+	return -EOPNOTSUPP;
+}
+static inline int cppc_get_epp_perf(int cpunum, u64 *epp_perf)
+{
+	return -EOPNOTSUPP;
+}
+static inline int cppc_set_epp(int cpu, u64 epp_val)
+{
+	return -EOPNOTSUPP;
+}
+static inline int cppc_get_auto_act_window(int cpu, u64 *auto_act_window)
+{
+	return -EOPNOTSUPP;
+}
+static inline int cppc_set_auto_act_window(int cpu, u64 auto_act_window)
+{
+	return -EOPNOTSUPP;
+}
+static inline int cppc_get_auto_sel(int cpu, bool *enable)
+{
+	return -EOPNOTSUPP;
+}
+static inline int cppc_set_auto_sel(int cpu, bool enable)
+{
+	return -EOPNOTSUPP;
+}
+static inline int cppc_get_perf_limited(int cpu, u64 *perf_limited)
+{
+	return -EOPNOTSUPP;
+}
+static inline int cppc_set_perf_limited(int cpu, u64 bits_to_clear)
+{
+	return -EOPNOTSUPP;
+}
+static inline int amd_get_highest_perf(unsigned int cpu, u32 *highest_perf)
+{
+	return -ENODEV;
+}
+static inline int amd_get_boost_ratio_numerator(unsigned int cpu, u64 *numerator)
+{
+	return -EOPNOTSUPP;
+}
+static inline int amd_detect_prefcore(bool *detected)
+{
+	return -ENODEV;
+}
+#endif /* !CONFIG_ACPI_CPPC_LIB */
+
+#endif /* _CPPC_ACPI_H*/