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, 60 insertions, 0 deletions

diff --git a/drivers/crypto/caam/dpseci-debugfs.c b/drivers/crypto/caam/dpseci-debugfs.c
new file mode 100644
index 000000000..020a9d8a8
--- /dev/null
+++ b/drivers/crypto/caam/dpseci-debugfs.c
@@ -0,0 +1,60 @@
+// SPDX-License-Identifier: (GPL-2.0+ OR BSD-3-Clause)
+/* Copyright 2019 NXP */
+
+#include <linux/module.h>
+#include <linux/device.h>
+#include <linux/debugfs.h>
+#include "dpseci-debugfs.h"
+
+static int dpseci_dbg_fqs_show(struct seq_file *file, void *offset)
+{
+	struct dpaa2_caam_priv *priv = file->private;
+	u32 fqid, fcnt, bcnt;
+	int i, err;
+
+	seq_printf(file, "FQ stats for %s:\n", dev_name(priv->dev));
+	seq_printf(file, "%s%16s%16s\n",
+		   "Rx-VFQID",
+		   "Pending frames",
+		   "Pending bytes");
+
+	for (i = 0; i <  priv->num_pairs; i++) {
+		fqid = priv->rx_queue_attr[i].fqid;
+		err = dpaa2_io_query_fq_count(NULL, fqid, &fcnt, &bcnt);
+		if (err)
+			continue;
+
+		seq_printf(file, "%5d%16u%16u\n", fqid, fcnt, bcnt);
+	}
+
+	seq_printf(file, "%s%16s%16s\n",
+		   "Tx-VFQID",
+		   "Pending frames",
+		   "Pending bytes");
+
+	for (i = 0; i <  priv->num_pairs; i++) {
+		fqid = priv->tx_queue_attr[i].fqid;
+		err = dpaa2_io_query_fq_count(NULL, fqid, &fcnt, &bcnt);
+		if (err)
+			continue;
+
+		seq_printf(file, "%5d%16u%16u\n", fqid, fcnt, bcnt);
+	}
+
+	return 0;
+}
+
+DEFINE_SHOW_ATTRIBUTE(dpseci_dbg_fqs);
+
+void dpaa2_dpseci_debugfs_init(struct dpaa2_caam_priv *priv)
+{
+	priv->dfs_root = debugfs_create_dir(dev_name(priv->dev), NULL);
+
+	debugfs_create_file("fq_stats", 0444, priv->dfs_root, priv,
+			    &dpseci_dbg_fqs_fops);
+}
+
+void dpaa2_dpseci_debugfs_exit(struct dpaa2_caam_priv *priv)
+{
+	debugfs_remove_recursive(priv->dfs_root);
+}