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, 87 insertions, 0 deletions

diff --git a/include/asm-generic/delay.h b/include/asm-generic/delay.h
new file mode 100644
index 000000000..03b0ec7af
--- /dev/null
+++ b/include/asm-generic/delay.h
@@ -0,0 +1,87 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __ASM_GENERIC_DELAY_H
+#define __ASM_GENERIC_DELAY_H
+
+#include <linux/math.h>
+#include <vdso/time64.h>
+
+/* Undefined functions to get compile-time errors */
+extern void __bad_udelay(void);
+extern void __bad_ndelay(void);
+
+extern void __udelay(unsigned long usecs);
+extern void __ndelay(unsigned long nsecs);
+extern void __const_udelay(unsigned long xloops);
+extern void __delay(unsigned long loops);
+
+/*
+ * The microseconds/nanosecond delay multiplicators are used to convert a
+ * constant microseconds/nanoseconds value to a value which can be used by the
+ * architectures specific implementation to transform it into loops.
+ */
+#define UDELAY_CONST_MULT	((unsigned long)DIV_ROUND_UP(1ULL << 32, USEC_PER_SEC))
+#define NDELAY_CONST_MULT	((unsigned long)DIV_ROUND_UP(1ULL << 32, NSEC_PER_SEC))
+
+/*
+ * The maximum constant udelay/ndelay value picked out of thin air to prevent
+ * too long constant udelays/ndelays.
+ */
+#define DELAY_CONST_MAX   20000
+
+/**
+ * udelay - Inserting a delay based on microseconds with busy waiting
+ * @usec:	requested delay in microseconds
+ *
+ * When delaying in an atomic context ndelay(), udelay() and mdelay() are the
+ * only valid variants of delaying/sleeping to go with.
+ *
+ * When inserting delays in non atomic context which are shorter than the time
+ * which is required to queue e.g. an hrtimer and to enter then the scheduler,
+ * it is also valuable to use udelay(). But it is not simple to specify a
+ * generic threshold for this which will fit for all systems. An approximation
+ * is a threshold for all delays up to 10 microseconds.
+ *
+ * When having a delay which is larger than the architecture specific
+ * %MAX_UDELAY_MS value, please make sure mdelay() is used. Otherwise a overflow
+ * risk is given.
+ *
+ * Please note that ndelay(), udelay() and mdelay() may return early for several
+ * reasons (https://lists.openwall.net/linux-kernel/2011/01/09/56):
+ *
+ * #. computed loops_per_jiffy too low (due to the time taken to execute the
+ *    timer interrupt.)
+ * #. cache behaviour affecting the time it takes to execute the loop function.
+ * #. CPU clock rate changes.
+ */
+static __always_inline void udelay(unsigned long usec)
+{
+	if (__builtin_constant_p(usec)) {
+		if (usec >= DELAY_CONST_MAX)
+			__bad_udelay();
+		else
+			__const_udelay(usec * UDELAY_CONST_MULT);
+	} else {
+		__udelay(usec);
+	}
+}
+
+/**
+ * ndelay - Inserting a delay based on nanoseconds with busy waiting
+ * @nsec:	requested delay in nanoseconds
+ *
+ * See udelay() for basic information about ndelay() and it's variants.
+ */
+static __always_inline void ndelay(unsigned long nsec)
+{
+	if (__builtin_constant_p(nsec)) {
+		if (nsec >= DELAY_CONST_MAX)
+			__bad_ndelay();
+		else
+			__const_udelay(nsec * NDELAY_CONST_MULT);
+	} else {
+		__ndelay(nsec);
+	}
+}
+#define ndelay(x) ndelay(x)
+
+#endif /* __ASM_GENERIC_DELAY_H */