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, 86 insertions, 0 deletions

diff --git a/include/crypto/internal/rsa.h b/include/crypto/internal/rsa.h
new file mode 100644
index 000000000..071a1951b
--- /dev/null
+++ b/include/crypto/internal/rsa.h
@@ -0,0 +1,86 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+/*
+ * RSA internal helpers
+ *
+ * Copyright (c) 2015, Intel Corporation
+ * Authors: Tadeusz Struk <tadeusz.struk@intel.com>
+ */
+#ifndef _RSA_HELPER_
+#define _RSA_HELPER_
+#include <linux/types.h>
+#include <crypto/akcipher.h>
+
+/**
+ * rsa_key - RSA key structure
+ * @n           : RSA modulus raw byte stream
+ * @e           : RSA public exponent raw byte stream
+ * @d           : RSA private exponent raw byte stream
+ * @p           : RSA prime factor p of n raw byte stream
+ * @q           : RSA prime factor q of n raw byte stream
+ * @dp          : RSA exponent d mod (p - 1) raw byte stream
+ * @dq          : RSA exponent d mod (q - 1) raw byte stream
+ * @qinv        : RSA CRT coefficient q^(-1) mod p raw byte stream
+ * @n_sz        : length in bytes of RSA modulus n
+ * @e_sz        : length in bytes of RSA public exponent
+ * @d_sz        : length in bytes of RSA private exponent
+ * @p_sz        : length in bytes of p field
+ * @q_sz        : length in bytes of q field
+ * @dp_sz       : length in bytes of dp field
+ * @dq_sz       : length in bytes of dq field
+ * @qinv_sz     : length in bytes of qinv field
+ */
+struct rsa_key {
+	const u8 *n;
+	const u8 *e;
+	const u8 *d;
+	const u8 *p;
+	const u8 *q;
+	const u8 *dp;
+	const u8 *dq;
+	const u8 *qinv;
+	size_t n_sz;
+	size_t e_sz;
+	size_t d_sz;
+	size_t p_sz;
+	size_t q_sz;
+	size_t dp_sz;
+	size_t dq_sz;
+	size_t qinv_sz;
+};
+
+int rsa_parse_pub_key(struct rsa_key *rsa_key, const void *key,
+		      unsigned int key_len);
+
+int rsa_parse_priv_key(struct rsa_key *rsa_key, const void *key,
+		       unsigned int key_len);
+
+#define RSA_PUB (true)
+#define RSA_PRIV (false)
+
+static inline int rsa_set_key(struct crypto_akcipher *child,
+			      unsigned int *key_size, bool is_pubkey,
+			      const void *key, unsigned int keylen)
+{
+	int err;
+
+	*key_size = 0;
+
+	if (is_pubkey)
+		err = crypto_akcipher_set_pub_key(child, key, keylen);
+	else
+		err = crypto_akcipher_set_priv_key(child, key, keylen);
+	if (err)
+		return err;
+
+	/* Find out new modulus size from rsa implementation */
+	err = crypto_akcipher_maxsize(child);
+	if (err > PAGE_SIZE)
+		return -ENOTSUPP;
+
+	*key_size = err;
+	return 0;
+}
+
+extern struct crypto_template rsa_pkcs1pad_tmpl;
+extern struct crypto_template rsassa_pkcs1_tmpl;
+#endif