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, 136 insertions, 0 deletions

diff --git a/arch/arm64/hyperv/mshyperv.c b/arch/arm64/hyperv/mshyperv.c
new file mode 100644
index 000000000..4fdc26ade
--- /dev/null
+++ b/arch/arm64/hyperv/mshyperv.c
@@ -0,0 +1,136 @@
+// SPDX-License-Identifier: GPL-2.0
+
+/*
+ * Core routines for interacting with Microsoft's Hyper-V hypervisor,
+ * including hypervisor initialization.
+ *
+ * Copyright (C) 2021, Microsoft, Inc.
+ *
+ * Author : Michael Kelley <mikelley@microsoft.com>
+ */
+
+#include <linux/types.h>
+#include <linux/acpi.h>
+#include <linux/export.h>
+#include <linux/errno.h>
+#include <linux/version.h>
+#include <linux/cpuhotplug.h>
+#include <asm/mshyperv.h>
+
+static bool hyperv_initialized;
+
+int hv_get_hypervisor_version(union hv_hypervisor_version_info *info)
+{
+	hv_get_vpreg_128(HV_REGISTER_HYPERVISOR_VERSION,
+			 (struct hv_get_vp_registers_output *)info);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(hv_get_hypervisor_version);
+
+#ifdef CONFIG_ACPI
+
+static bool __init hyperv_detect_via_acpi(void)
+{
+	if (acpi_disabled)
+		return false;
+	/*
+	 * Hypervisor ID is only available in ACPI v6+, and the
+	 * structure layout was extended in v6 to accommodate that
+	 * new field.
+	 *
+	 * At the very minimum, this check makes sure not to read
+	 * past the FADT structure.
+	 *
+	 * It is also needed to catch running in some unknown
+	 * non-Hyper-V environment that has ACPI 5.x or less.
+	 * In such a case, it can't be Hyper-V.
+	 */
+	if (acpi_gbl_FADT.header.revision < 6)
+		return false;
+	return strncmp((char *)&acpi_gbl_FADT.hypervisor_id, "MsHyperV", 8) == 0;
+}
+
+#else
+
+static bool __init hyperv_detect_via_acpi(void)
+{
+	return false;
+}
+
+#endif
+
+static bool __init hyperv_detect_via_smccc(void)
+{
+	uuid_t hyperv_uuid = UUID_INIT(
+		0x58ba324d, 0x6447, 0x24cd,
+		0x75, 0x6c, 0xef, 0x8e,
+		0x24, 0x70, 0x59, 0x16);
+
+	return arm_smccc_hypervisor_has_uuid(&hyperv_uuid);
+}
+
+static int __init hyperv_init(void)
+{
+	struct hv_get_vp_registers_output	result;
+	u64	guest_id;
+	int	ret;
+
+	/*
+	 * Allow for a kernel built with CONFIG_HYPERV to be running in
+	 * a non-Hyper-V environment.
+	 *
+	 * In such cases, do nothing and return success.
+	 */
+	if (!hyperv_detect_via_acpi() && !hyperv_detect_via_smccc())
+		return 0;
+
+	/* Setup the guest ID */
+	guest_id = hv_generate_guest_id(LINUX_VERSION_CODE);
+	hv_set_vpreg(HV_REGISTER_GUEST_OS_ID, guest_id);
+
+	/* Get the features and hints from Hyper-V */
+	hv_get_vpreg_128(HV_REGISTER_PRIVILEGES_AND_FEATURES_INFO, &result);
+	ms_hyperv.features = result.as32.a;
+	ms_hyperv.priv_high = result.as32.b;
+	ms_hyperv.misc_features = result.as32.c;
+
+	hv_get_vpreg_128(HV_REGISTER_FEATURES_INFO, &result);
+	ms_hyperv.hints = result.as32.a;
+
+	pr_info("Hyper-V: privilege flags low 0x%x, high 0x%x, hints 0x%x, misc 0x%x\n",
+		ms_hyperv.features, ms_hyperv.priv_high, ms_hyperv.hints,
+		ms_hyperv.misc_features);
+
+	hv_identify_partition_type();
+
+	ret = hv_common_init();
+	if (ret)
+		return ret;
+
+	ret = cpuhp_setup_state(CPUHP_AP_HYPERV_ONLINE, "arm64/hyperv_init:online",
+				hv_common_cpu_init, hv_common_cpu_die);
+	if (ret < 0) {
+		hv_common_free();
+		return ret;
+	}
+
+	if (ms_hyperv.priv_high & HV_ACCESS_PARTITION_ID)
+		hv_get_partition_id();
+	ms_hyperv.vtl = get_vtl();
+	if (ms_hyperv.vtl > 0) /* non default VTL */
+		pr_info("Linux runs in Hyper-V Virtual Trust Level %d\n", ms_hyperv.vtl);
+
+	ms_hyperv_late_init();
+
+	hyperv_initialized = true;
+	return 0;
+}
+
+early_initcall(hyperv_init);
+
+bool hv_is_hyperv_initialized(void)
+{
+	return hyperv_initialized;
+}
+EXPORT_SYMBOL_GPL(hv_is_hyperv_initialized);