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, 98 insertions, 0 deletions

diff --git a/scripts/include/hashtable.h b/scripts/include/hashtable.h
new file mode 100644
index 000000000..45abcb12b
--- /dev/null
+++ b/scripts/include/hashtable.h
@@ -0,0 +1,98 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+#ifndef HASHTABLE_H
+#define HASHTABLE_H
+
+#include "array_size.h"
+#include "list.h"
+
+#define HASH_SIZE(name) (ARRAY_SIZE(name))
+
+#define HASHTABLE_DECLARE(name, size)		struct hlist_head name[size]
+
+#define HASHTABLE_DEFINE(name, size)						\
+	HASHTABLE_DECLARE(name, size) =						\
+			{ [0 ... ((size) - 1)] = HLIST_HEAD_INIT }
+
+#define hash_head(table, key)		(&(table)[(key) % HASH_SIZE(table)])
+
+static inline void __hash_init(struct hlist_head *ht, unsigned int sz)
+{
+	unsigned int i;
+
+	for (i = 0; i < sz; i++)
+		INIT_HLIST_HEAD(&ht[i]);
+}
+
+/**
+ * hash_init - initialize a hash table
+ * @table: hashtable to be initialized
+ *
+ * This has to be a macro since HASH_SIZE() will not work on pointers since
+ * it calculates the size during preprocessing.
+ */
+#define hash_init(table) __hash_init(table, HASH_SIZE(table))
+
+/**
+ * hash_add - add an object to a hashtable
+ * @table: hashtable to add to
+ * @node: the &struct hlist_node of the object to be added
+ * @key: the key of the object to be added
+ */
+#define hash_add(table, node, key)						\
+	hlist_add_head(node, hash_head(table, key))
+
+/**
+ * hash_del - remove an object from a hashtable
+ * @node: &struct hlist_node of the object to remove
+ */
+static inline void hash_del(struct hlist_node *node)
+{
+	hlist_del_init(node);
+}
+
+/**
+ * hash_for_each - iterate over a hashtable
+ * @table: hashtable to iterate
+ * @obj: the type * to use as a loop cursor for each entry
+ * @member: the name of the hlist_node within the struct
+ */
+#define hash_for_each(table, obj, member)				\
+	for (int _bkt = 0; _bkt < HASH_SIZE(table); _bkt++)		\
+		hlist_for_each_entry(obj, &table[_bkt], member)
+
+/**
+ * hash_for_each_safe - iterate over a hashtable safe against removal of
+ * hash entry
+ * @table: hashtable to iterate
+ * @obj: the type * to use as a loop cursor for each entry
+ * @tmp: a &struct hlist_node used for temporary storage
+ * @member: the name of the hlist_node within the struct
+ */
+#define hash_for_each_safe(table, obj, tmp, member)			\
+	for (int _bkt = 0; _bkt < HASH_SIZE(table); _bkt++)		\
+		hlist_for_each_entry_safe(obj, tmp, &table[_bkt], member)
+
+/**
+ * hash_for_each_possible - iterate over all possible objects hashing to the
+ * same bucket
+ * @table: hashtable to iterate
+ * @obj: the type * to use as a loop cursor for each entry
+ * @member: the name of the hlist_node within the struct
+ * @key: the key of the objects to iterate over
+ */
+#define hash_for_each_possible(table, obj, member, key)			\
+	hlist_for_each_entry(obj, hash_head(table, key), member)
+
+/**
+ * hash_for_each_possible_safe - iterate over all possible objects hashing to the
+ * same bucket safe against removals
+ * @table: hashtable to iterate
+ * @obj: the type * to use as a loop cursor for each entry
+ * @tmp: a &struct hlist_node used for temporary storage
+ * @member: the name of the hlist_node within the struct
+ * @key: the key of the objects to iterate over
+ */
+#define hash_for_each_possible_safe(table, obj, tmp, member, key)	\
+	hlist_for_each_entry_safe(obj, tmp, hash_head(table, key), member)
+
+#endif /* HASHTABLE_H */