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, 42 insertions, 0 deletions

diff --git a/include/asm-generic/msi.h b/include/asm-generic/msi.h
new file mode 100644
index 000000000..92cca4b23
--- /dev/null
+++ b/include/asm-generic/msi.h
@@ -0,0 +1,42 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __ASM_GENERIC_MSI_H
+#define __ASM_GENERIC_MSI_H
+
+#include <linux/types.h>
+
+#ifdef CONFIG_GENERIC_MSI_IRQ
+
+#ifndef NUM_MSI_ALLOC_SCRATCHPAD_REGS
+# define NUM_MSI_ALLOC_SCRATCHPAD_REGS	2
+#endif
+
+struct msi_desc;
+
+/**
+ * struct msi_alloc_info - Default structure for MSI interrupt allocation.
+ * @desc:	Pointer to msi descriptor
+ * @hwirq:	Associated hw interrupt number in the domain
+ * @scratchpad:	Storage for implementation specific scratch data
+ *
+ * Architectures can provide their own implementation by not including
+ * asm-generic/msi.h into their arch specific header file.
+ */
+typedef struct msi_alloc_info {
+	struct msi_desc			*desc;
+	irq_hw_number_t			hwirq;
+	unsigned long			flags;
+	union {
+		unsigned long		ul;
+		void			*ptr;
+	} scratchpad[NUM_MSI_ALLOC_SCRATCHPAD_REGS];
+} msi_alloc_info_t;
+
+/* Device generating MSIs is proxying for another device */
+#define MSI_ALLOC_FLAGS_PROXY_DEVICE	(1UL << 0)
+#define MSI_ALLOC_FLAGS_FIXED_MSG_DATA	(1UL << 1)
+
+#define GENERIC_MSI_DOMAIN_OPS		1
+
+#endif /* CONFIG_GENERIC_MSI_IRQ */
+
+#endif