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, 56 insertions, 0 deletions

diff --git a/arch/powerpc/sysdev/ipic.h b/arch/powerpc/sysdev/ipic.h
new file mode 100644
index 000000000..45ab61406
--- /dev/null
+++ b/arch/powerpc/sysdev/ipic.h
@@ -0,0 +1,56 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * IPIC private definitions and structure.
+ *
+ * Maintainer: Kumar Gala <galak@kernel.crashing.org>
+ *
+ * Copyright 2005 Freescale Semiconductor, Inc
+ */
+#ifndef __IPIC_H__
+#define __IPIC_H__
+
+#include <asm/ipic.h>
+
+#define NR_IPIC_INTS 128
+
+/* External IRQS */
+#define IPIC_IRQ_EXT0 48
+#define IPIC_IRQ_EXT1 17
+#define IPIC_IRQ_EXT7 23
+
+/* Default Priority Registers */
+#define IPIC_PRIORITY_DEFAULT 0x05309770
+
+/* System Global Interrupt Configuration Register */
+#define	SICFR_IPSA	0x00010000
+#define	SICFR_IPSB	0x00020000
+#define	SICFR_IPSC	0x00040000
+#define	SICFR_IPSD	0x00080000
+#define	SICFR_MPSA	0x00200000
+#define	SICFR_MPSB	0x00400000
+
+/* System External Interrupt Mask Register */
+#define	SEMSR_SIRQ0	0x00008000
+
+/* System Error Control Register */
+#define SERCR_MCPR	0x00000001
+
+struct ipic {
+	volatile u32 __iomem	*regs;
+
+	/* The remapper for this IPIC */
+	struct irq_domain		*irqhost;
+};
+
+struct ipic_info {
+	u8	ack;		/* pending register offset from base if the irq
+				   supports ack operation */
+	u8	mask;		/* mask register offset from base */
+	u8	prio;		/* priority register offset from base */
+	u8	force;		/* force register offset from base */
+	u8	bit;		/* register bit position (as per doc)
+				   bit mask = 1 << (31 - bit) */
+	u8	prio_mask;	/* priority mask value */
+};
+
+#endif /* __IPIC_H__ */