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, 52 insertions, 0 deletions

diff --git a/include/asm-generic/getorder.h b/include/asm-generic/getorder.h
new file mode 100644
index 000000000..875ccae19
--- /dev/null
+++ b/include/asm-generic/getorder.h
@@ -0,0 +1,52 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __ASM_GENERIC_GETORDER_H
+#define __ASM_GENERIC_GETORDER_H
+
+#ifndef __ASSEMBLER__
+
+#include <linux/compiler.h>
+#include <linux/log2.h>
+
+/**
+ * get_order - Determine the allocation order of a memory size
+ * @size: The size for which to get the order
+ *
+ * Determine the allocation order of a particular sized block of memory.  This
+ * is on a logarithmic scale, where:
+ *
+ *	0 -> 2^0 * PAGE_SIZE and below
+ *	1 -> 2^1 * PAGE_SIZE to 2^0 * PAGE_SIZE + 1
+ *	2 -> 2^2 * PAGE_SIZE to 2^1 * PAGE_SIZE + 1
+ *	3 -> 2^3 * PAGE_SIZE to 2^2 * PAGE_SIZE + 1
+ *	4 -> 2^4 * PAGE_SIZE to 2^3 * PAGE_SIZE + 1
+ *	...
+ *
+ * The order returned is used to find the smallest allocation granule required
+ * to hold an object of the specified size.
+ *
+ * The result is undefined if the size is 0.
+ */
+static __always_inline __attribute_const__ int get_order(unsigned long size)
+{
+	if (__builtin_constant_p(size)) {
+		if (!size)
+			return BITS_PER_LONG - PAGE_SHIFT;
+
+		if (size < (1UL << PAGE_SHIFT))
+			return 0;
+
+		return ilog2((size) - 1) - PAGE_SHIFT + 1;
+	}
+
+	size--;
+	size >>= PAGE_SHIFT;
+#if BITS_PER_LONG == 32
+	return fls(size);
+#else
+	return fls64(size);
+#endif
+}
+
+#endif	/* __ASSEMBLER__ */
+
+#endif	/* __ASM_GENERIC_GETORDER_H */