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, 101 insertions, 0 deletions

diff --git a/net/netlabel/netlabel_user.c b/net/netlabel/netlabel_user.c
new file mode 100644
index 000000000..0da652844
--- /dev/null
+++ b/net/netlabel/netlabel_user.c
@@ -0,0 +1,101 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * NetLabel NETLINK Interface
+ *
+ * This file defines the NETLINK interface for the NetLabel system.  The
+ * NetLabel system manages static and dynamic label mappings for network
+ * protocols such as CIPSO and RIPSO.
+ *
+ * Author: Paul Moore <paul@paul-moore.com>
+ */
+
+/*
+ * (c) Copyright Hewlett-Packard Development Company, L.P., 2006
+ */
+
+#include <linux/init.h>
+#include <linux/types.h>
+#include <linux/list.h>
+#include <linux/socket.h>
+#include <linux/audit.h>
+#include <linux/tty.h>
+#include <linux/security.h>
+#include <linux/gfp.h>
+#include <net/sock.h>
+#include <net/netlink.h>
+#include <net/genetlink.h>
+#include <net/netlabel.h>
+#include <asm/bug.h>
+
+#include "netlabel_mgmt.h"
+#include "netlabel_unlabeled.h"
+#include "netlabel_cipso_v4.h"
+#include "netlabel_calipso.h"
+#include "netlabel_user.h"
+
+/*
+ * NetLabel NETLINK Setup Functions
+ */
+
+/**
+ * netlbl_netlink_init - Initialize the NETLINK communication channel
+ *
+ * Description:
+ * Call out to the NetLabel components so they can register their families and
+ * commands with the Generic NETLINK mechanism.  Returns zero on success and
+ * non-zero on failure.
+ *
+ */
+int __init netlbl_netlink_init(void)
+{
+	int ret_val;
+
+	ret_val = netlbl_mgmt_genl_init();
+	if (ret_val != 0)
+		return ret_val;
+
+	ret_val = netlbl_cipsov4_genl_init();
+	if (ret_val != 0)
+		return ret_val;
+
+	ret_val = netlbl_calipso_genl_init();
+	if (ret_val != 0)
+		return ret_val;
+
+	return netlbl_unlabel_genl_init();
+}
+
+/*
+ * NetLabel Audit Functions
+ */
+
+/**
+ * netlbl_audit_start_common - Start an audit message
+ * @type: audit message type
+ * @audit_info: NetLabel audit information
+ *
+ * Description:
+ * Start an audit message using the type specified in @type and fill the audit
+ * message with some fields common to all NetLabel audit messages.  Returns
+ * a pointer to the audit buffer on success, NULL on failure.
+ *
+ */
+struct audit_buffer *netlbl_audit_start_common(int type,
+					       struct netlbl_audit *audit_info)
+{
+	struct audit_buffer *audit_buf;
+
+	if (audit_enabled == AUDIT_OFF)
+		return NULL;
+
+	audit_buf = audit_log_start(audit_context(), GFP_ATOMIC, type);
+	if (audit_buf == NULL)
+		return NULL;
+
+	audit_log_format(audit_buf, "netlabel: auid=%u ses=%u",
+			 from_kuid(&init_user_ns, audit_info->loginuid),
+			 audit_info->sessionid);
+	audit_log_subj_ctx(audit_buf, &audit_info->prop);
+
+	return audit_buf;
+}