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, 116 insertions, 0 deletions

diff --git a/tools/perf/util/tool.h b/tools/perf/util/tool.h
new file mode 100644
index 000000000..2a4f124ff
--- /dev/null
+++ b/tools/perf/util/tool.h
@@ -0,0 +1,116 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __PERF_TOOL_H
+#define __PERF_TOOL_H
+
+#include <stdbool.h>
+
+#include <linux/types.h>
+
+struct perf_session;
+union perf_event;
+struct evlist;
+struct perf_sample;
+struct perf_tool;
+struct machine;
+struct ordered_events;
+
+typedef int (*event_sample)(const struct perf_tool *tool, union perf_event *event,
+			    struct perf_sample *sample,
+			    struct machine *machine);
+
+typedef int (*event_op)(const struct perf_tool *tool, union perf_event *event,
+			struct perf_sample *sample, struct machine *machine);
+
+typedef int (*event_attr_op)(const struct perf_tool *tool,
+			     union perf_event *event,
+			     struct evlist **pevlist);
+
+typedef int (*event_op2)(const struct perf_tool *tool, struct perf_session *session,
+			 union perf_event *event);
+typedef s64 (*event_op3)(const struct perf_tool *tool, struct perf_session *session,
+			 union perf_event *event);
+typedef int (*event_op4)(const struct perf_tool *tool, struct perf_session *session,
+			 union perf_event *event, u64 data, const char *str);
+
+typedef int (*event_oe)(const struct perf_tool *tool, union perf_event *event,
+			struct ordered_events *oe);
+
+enum show_feature_header {
+	SHOW_FEAT_NO_HEADER = 0,
+	SHOW_FEAT_HEADER,
+	SHOW_FEAT_HEADER_FULL_INFO,
+};
+
+struct perf_tool {
+	event_sample	sample,
+			read,
+			callchain_deferred;
+	event_op	mmap,
+			mmap2,
+			comm,
+			namespaces,
+			cgroup,
+			fork,
+			exit,
+			lost,
+			lost_samples,
+			aux,
+			itrace_start,
+			aux_output_hw_id,
+			context_switch,
+			throttle,
+			unthrottle,
+			ksymbol,
+			bpf,
+			text_poke;
+
+	event_attr_op	attr;
+	event_attr_op	event_update;
+	event_op2	tracing_data;
+	event_oe	finished_round;
+	event_op2	build_id,
+			id_index,
+			auxtrace_info,
+			auxtrace_error,
+			time_conv,
+			thread_map,
+			cpu_map,
+			stat_config,
+			stat,
+			stat_round,
+			feature,
+			finished_init,
+			bpf_metadata,
+			schedstat_cpu,
+			schedstat_domain;
+	event_op4	compressed;
+	event_op3	auxtrace;
+	bool		ordered_events;
+	bool		ordering_requires_timestamps;
+	bool		namespace_events;
+	bool		cgroup_events;
+	bool		no_warn;
+	bool		dont_split_sample_group;
+	bool		merge_deferred_callchains;
+	enum show_feature_header show_feat_hdr;
+};
+
+void perf_tool__init(struct perf_tool *tool, bool ordered_events);
+
+bool perf_tool__compressed_is_stub(const struct perf_tool *tool);
+
+int process_event_sample_stub(const struct perf_tool *tool,
+			      union perf_event *event,
+			      struct perf_sample *sample,
+			      struct machine *machine);
+
+struct delegate_tool {
+	/** @tool: The actual tool that calls the delegate. */
+	struct perf_tool tool;
+	/** @delegate: The tool that is delegated to. */
+	struct perf_tool *delegate;
+};
+
+void delegate_tool__init(struct delegate_tool *tool, struct perf_tool *delegate);
+
+#endif /* __PERF_TOOL_H */