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, 106 insertions, 0 deletions

diff --git a/crypto/xxhash_generic.c b/crypto/xxhash_generic.c
new file mode 100644
index 000000000..175bb7ae0
--- /dev/null
+++ b/crypto/xxhash_generic.c
@@ -0,0 +1,106 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#include <crypto/internal/hash.h>
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/xxhash.h>
+#include <linux/unaligned.h>
+
+#define XXHASH64_BLOCK_SIZE	32
+#define XXHASH64_DIGEST_SIZE	8
+
+struct xxhash64_tfm_ctx {
+	u64 seed;
+};
+
+struct xxhash64_desc_ctx {
+	struct xxh64_state xxhstate;
+};
+
+static int xxhash64_setkey(struct crypto_shash *tfm, const u8 *key,
+			 unsigned int keylen)
+{
+	struct xxhash64_tfm_ctx *tctx = crypto_shash_ctx(tfm);
+
+	if (keylen != sizeof(tctx->seed))
+		return -EINVAL;
+	tctx->seed = get_unaligned_le64(key);
+	return 0;
+}
+
+static int xxhash64_init(struct shash_desc *desc)
+{
+	struct xxhash64_tfm_ctx *tctx = crypto_shash_ctx(desc->tfm);
+	struct xxhash64_desc_ctx *dctx = shash_desc_ctx(desc);
+
+	xxh64_reset(&dctx->xxhstate, tctx->seed);
+
+	return 0;
+}
+
+static int xxhash64_update(struct shash_desc *desc, const u8 *data,
+			 unsigned int length)
+{
+	struct xxhash64_desc_ctx *dctx = shash_desc_ctx(desc);
+
+	xxh64_update(&dctx->xxhstate, data, length);
+
+	return 0;
+}
+
+static int xxhash64_final(struct shash_desc *desc, u8 *out)
+{
+	struct xxhash64_desc_ctx *dctx = shash_desc_ctx(desc);
+
+	put_unaligned_le64(xxh64_digest(&dctx->xxhstate), out);
+
+	return 0;
+}
+
+static int xxhash64_digest(struct shash_desc *desc, const u8 *data,
+			 unsigned int length, u8 *out)
+{
+	struct xxhash64_tfm_ctx *tctx = crypto_shash_ctx(desc->tfm);
+
+	put_unaligned_le64(xxh64(data, length, tctx->seed), out);
+
+	return 0;
+}
+
+static struct shash_alg alg = {
+	.digestsize	= XXHASH64_DIGEST_SIZE,
+	.setkey		= xxhash64_setkey,
+	.init		= xxhash64_init,
+	.update		= xxhash64_update,
+	.final		= xxhash64_final,
+	.digest		= xxhash64_digest,
+	.descsize	= sizeof(struct xxhash64_desc_ctx),
+	.base		= {
+		.cra_name	 = "xxhash64",
+		.cra_driver_name = "xxhash64-generic",
+		.cra_priority	 = 100,
+		.cra_flags	 = CRYPTO_ALG_OPTIONAL_KEY,
+		.cra_blocksize	 = XXHASH64_BLOCK_SIZE,
+		.cra_ctxsize	 = sizeof(struct xxhash64_tfm_ctx),
+		.cra_module	 = THIS_MODULE,
+	}
+};
+
+static int __init xxhash_mod_init(void)
+{
+	return crypto_register_shash(&alg);
+}
+
+static void __exit xxhash_mod_fini(void)
+{
+	crypto_unregister_shash(&alg);
+}
+
+module_init(xxhash_mod_init);
+module_exit(xxhash_mod_fini);
+
+MODULE_AUTHOR("Nikolay Borisov <nborisov@suse.com>");
+MODULE_DESCRIPTION("xxhash calculations wrapper for lib/xxhash.c");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS_CRYPTO("xxhash64");
+MODULE_ALIAS_CRYPTO("xxhash64-generic");