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, 92 insertions, 0 deletions

diff --git a/drivers/usb/chipidea/trace.h b/drivers/usb/chipidea/trace.h
new file mode 100644
index 000000000..1875419cd
--- /dev/null
+++ b/drivers/usb/chipidea/trace.h
@@ -0,0 +1,92 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Trace support header file for device mode
+ *
+ * Copyright (C) 2020 NXP
+ *
+ * Author: Peter Chen <peter.chen@nxp.com>
+ */
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM chipidea
+
+#if !defined(__LINUX_CHIPIDEA_TRACE) || defined(TRACE_HEADER_MULTI_READ)
+#define __LINUX_CHIPIDEA_TRACE
+
+#include <linux/types.h>
+#include <linux/tracepoint.h>
+#include <linux/usb/chipidea.h>
+#include "ci.h"
+#include "udc.h"
+
+#define CHIPIDEA_MSG_MAX	500
+
+void ci_log(struct ci_hdrc *ci, const char *fmt, ...);
+
+TRACE_EVENT(ci_log,
+	TP_PROTO(struct ci_hdrc *ci, struct va_format *vaf),
+	TP_ARGS(ci, vaf),
+	TP_STRUCT__entry(
+		__string(name, dev_name(ci->dev))
+		__vstring(msg, vaf->fmt, vaf->va)
+	),
+	TP_fast_assign(
+		__assign_str(name);
+		__assign_vstr(msg, vaf->fmt, vaf->va);
+	),
+	TP_printk("%s: %s", __get_str(name), __get_str(msg))
+);
+
+DECLARE_EVENT_CLASS(ci_log_trb,
+	TP_PROTO(struct ci_hw_ep *hwep, struct ci_hw_req *hwreq, struct td_node *td),
+	TP_ARGS(hwep, hwreq, td),
+	TP_STRUCT__entry(
+		__string(name, hwep->name)
+		__field(struct td_node *, td)
+		__field(struct usb_request *, req)
+		__field(dma_addr_t, dma)
+		__field(s32, td_remaining_size)
+		__field(u32, next)
+		__field(u32, token)
+		__field(u32, type)
+	),
+	TP_fast_assign(
+		__assign_str(name);
+		__entry->req = &hwreq->req;
+		__entry->td = td;
+		__entry->dma = td->dma;
+		__entry->td_remaining_size = td->td_remaining_size;
+		__entry->next = le32_to_cpu(td->ptr->next);
+		__entry->token = le32_to_cpu(td->ptr->token);
+		__entry->type = usb_endpoint_type(hwep->ep.desc);
+	),
+	TP_printk("%s: req: %p, td: %p, td_dma_address: %pad, remaining_size: %d, "
+	       "next: %x, total bytes: %d, status: %lx",
+		__get_str(name), __entry->req, __entry->td, &__entry->dma,
+		__entry->td_remaining_size, __entry->next,
+		(int)((__entry->token & TD_TOTAL_BYTES) >> __ffs(TD_TOTAL_BYTES)),
+		__entry->token & TD_STATUS
+	)
+);
+
+DEFINE_EVENT(ci_log_trb, ci_prepare_td,
+	TP_PROTO(struct ci_hw_ep *hwep, struct ci_hw_req *hwreq, struct td_node *td),
+	TP_ARGS(hwep, hwreq, td)
+);
+
+DEFINE_EVENT(ci_log_trb, ci_complete_td,
+	TP_PROTO(struct ci_hw_ep *hwep, struct ci_hw_req *hwreq, struct td_node *td),
+	TP_ARGS(hwep, hwreq, td)
+);
+
+#endif /* __LINUX_CHIPIDEA_TRACE */
+
+/* this part must be outside header guard */
+
+#undef TRACE_INCLUDE_PATH
+#define TRACE_INCLUDE_PATH .
+
+#undef TRACE_INCLUDE_FILE
+#define TRACE_INCLUDE_FILE trace
+
+#include <trace/define_trace.h>