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, 69 insertions, 0 deletions

diff --git a/net/rxrpc/call_state.c b/net/rxrpc/call_state.c
new file mode 100644
index 000000000..6afb54373
--- /dev/null
+++ b/net/rxrpc/call_state.c
@@ -0,0 +1,69 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/* Call state changing functions.
+ *
+ * Copyright (C) 2022 Red Hat, Inc. All Rights Reserved.
+ * Written by David Howells (dhowells@redhat.com)
+ */
+
+#include "ar-internal.h"
+
+/*
+ * Transition a call to the complete state.
+ */
+bool rxrpc_set_call_completion(struct rxrpc_call *call,
+				 enum rxrpc_call_completion compl,
+				 u32 abort_code,
+				 int error)
+{
+	if (__rxrpc_call_state(call) == RXRPC_CALL_COMPLETE)
+		return false;
+
+	call->abort_code = abort_code;
+	call->error = error;
+	call->completion = compl;
+	/* Allow reader of completion state to operate locklessly */
+	rxrpc_set_call_state(call, RXRPC_CALL_COMPLETE);
+	trace_rxrpc_call_complete(call);
+	wake_up(&call->waitq);
+	rxrpc_notify_socket(call);
+	return true;
+}
+
+/*
+ * Record that a call successfully completed.
+ */
+bool rxrpc_call_completed(struct rxrpc_call *call)
+{
+	return rxrpc_set_call_completion(call, RXRPC_CALL_SUCCEEDED, 0, 0);
+}
+
+/*
+ * Record that a call is locally aborted.
+ */
+bool rxrpc_abort_call(struct rxrpc_call *call, rxrpc_seq_t seq,
+		      u32 abort_code, int error, enum rxrpc_abort_reason why)
+{
+	trace_rxrpc_abort(call->debug_id, why, call->cid, call->call_id, seq,
+			  abort_code, error);
+	if (!rxrpc_set_call_completion(call, RXRPC_CALL_LOCALLY_ABORTED,
+				       abort_code, error))
+		return false;
+	if (test_bit(RXRPC_CALL_EXPOSED, &call->flags))
+		rxrpc_send_abort_packet(call);
+	return true;
+}
+
+/*
+ * Record that a call errored out before even getting off the ground, thereby
+ * setting the state to allow it to be destroyed.
+ */
+void rxrpc_prefail_call(struct rxrpc_call *call, enum rxrpc_call_completion compl,
+			int error)
+{
+	call->abort_code	= RX_CALL_DEAD;
+	call->error		= error;
+	call->completion	= compl;
+	call->_state		= RXRPC_CALL_COMPLETE;
+	trace_rxrpc_call_complete(call);
+	WARN_ON_ONCE(__test_and_set_bit(RXRPC_CALL_RELEASED, &call->flags));
+}