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, 139 insertions, 0 deletions

diff --git a/net/mac80211/drop.h b/net/mac80211/drop.h
new file mode 100644
index 000000000..f06a8aa90
--- /dev/null
+++ b/net/mac80211/drop.h
@@ -0,0 +1,139 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * mac80211 drop reason list
+ *
+ * Copyright (C) 2023-2024, 2026 Intel Corporation
+ */
+
+#ifndef MAC80211_DROP_H
+#define MAC80211_DROP_H
+#include <net/dropreason.h>
+
+typedef unsigned int __bitwise ieee80211_rx_result;
+
+#define MAC80211_DROP_REASONS_UNUSABLE(R)	\
+	/* 0x00 == ___RX_DROP_UNUSABLE */	\
+	R(RX_DROP_U_MIC_FAIL)			\
+	R(RX_DROP_U_REPLAY)			\
+	R(RX_DROP_U_BAD_MMIE)			\
+	R(RX_DROP_U_DUP)			\
+	R(RX_DROP_U_SPURIOUS)			\
+	R(RX_DROP_U_DECRYPT_FAIL)		\
+	R(RX_DROP_U_NO_KEY_ID)			\
+	R(RX_DROP_U_BAD_CIPHER)			\
+	R(RX_DROP_U_OOM)			\
+	R(RX_DROP_U_NONSEQ_PN)			\
+	R(RX_DROP_U_BAD_KEY_COLOR)		\
+	R(RX_DROP_U_BAD_4ADDR)			\
+	R(RX_DROP_U_BAD_AMSDU)			\
+	R(RX_DROP_U_BAD_AMSDU_CIPHER)		\
+	R(RX_DROP_U_INVALID_8023)		\
+	/* 0x10 */				\
+	R(RX_DROP_U_RUNT_ACTION)		\
+	R(RX_DROP_U_UNPROT_ACTION)		\
+	R(RX_DROP_U_UNPROT_DUAL)		\
+	R(RX_DROP_U_UNPROT_UCAST_MGMT)		\
+	R(RX_DROP_U_UNPROT_MCAST_MGMT)		\
+	R(RX_DROP_U_UNPROT_BEACON)		\
+	R(RX_DROP_U_UNPROT_UNICAST_PUB_ACTION)	\
+	R(RX_DROP_U_UNPROT_ROBUST_ACTION)	\
+	R(RX_DROP_U_ACTION_UNKNOWN_SRC)		\
+	R(RX_DROP_U_REJECTED_ACTION_RESPONSE)	\
+	R(RX_DROP_U_EXPECT_DEFRAG_PROT)		\
+	R(RX_DROP_U_WEP_DEC_FAIL)		\
+	R(RX_DROP_U_NO_IV)			\
+	R(RX_DROP_U_NO_ICV)			\
+	R(RX_DROP_U_AP_RX_GROUPCAST)		\
+	R(RX_DROP_U_SHORT_MMIC)			\
+	/* 0x20 */				\
+	R(RX_DROP_U_MMIC_FAIL)			\
+	R(RX_DROP_U_SHORT_TKIP)			\
+	R(RX_DROP_U_TKIP_FAIL)			\
+	R(RX_DROP_U_SHORT_CCMP)			\
+	R(RX_DROP_U_SHORT_CCMP_MIC)		\
+	R(RX_DROP_U_SHORT_GCMP)			\
+	R(RX_DROP_U_SHORT_GCMP_MIC)		\
+	R(RX_DROP_U_SHORT_CMAC)			\
+	R(RX_DROP_U_SHORT_CMAC256)		\
+	R(RX_DROP_U_SHORT_GMAC)			\
+	R(RX_DROP_U_UNEXPECTED_VLAN_4ADDR)	\
+	R(RX_DROP_U_UNEXPECTED_STA_4ADDR)	\
+	R(RX_DROP_U_UNEXPECTED_VLAN_MCAST)	\
+	R(RX_DROP_U_NOT_PORT_CONTROL)		\
+	R(RX_DROP_U_UNEXPECTED_4ADDR_FRAME)	\
+	R(RX_DROP_U_BAD_BCN_KEYIDX)		\
+	/* 0x30 */				\
+	R(RX_DROP_U_BAD_MGMT_KEYIDX)		\
+	R(RX_DROP_U_UNKNOWN_ACTION_REJECTED)	\
+	R(RX_DROP_U_MESH_DS_BITS)		\
+	R(RX_DROP_U_MESH_A3_MISMATCH)		\
+	R(RX_DROP_U_MESH_NO_A4)			\
+	R(RX_DROP_U_MESH_A4_MISMATCH)		\
+	R(RX_DROP_U_MESH_UNEXP_DATA)		\
+	R(RX_DROP_U_MESH_WRONG_ACTION)		\
+	R(RX_DROP_U_MESH_UNEXP_MGMT)		\
+	R(RX_DROP_U_SPURIOUS_NOTIF)		\
+	R(RX_DROP_U_RUNT_DATA)			\
+	R(RX_DROP_U_KEY_TAINTED)		\
+	R(RX_DROP_U_UNPROTECTED)		\
+	R(RX_DROP_U_MCAST_FRAGMENT)		\
+	R(RX_DROP_U_DEFRAG_MISMATCH)		\
+	R(RX_DROP_U_RUNT_MESH_DATA)		\
+	/* 0x40 */				\
+	R(RX_DROP_U_MESH_NO_TTL)		\
+	R(RX_DROP_U_MESH_RMC)			\
+	R(RX_DROP_U_MESH_BAD_AE)		\
+	R(RX_DROP_U_MESH_TTL_EXPIRED)		\
+	R(RX_DROP_U_MESH_NOT_FORWARDING)	\
+	R(RX_DROP_U_AMSDU_WITHOUT_DATA)		\
+	R(RX_DROP_U_NULL_DATA)			\
+	R(RX_DROP_U_UNEXPECTED_4ADDR)		\
+	R(RX_DROP_U_PORT_CONTROL)		\
+	R(RX_DROP_U_UNKNOWN_STA)		\
+	R(RX_DROP_U_RUNT_BAR)			\
+	R(RX_DROP_U_BAR_OUTSIDE_SESSION)	\
+	R(RX_DROP_U_CTRL_FRAME)			\
+	R(RX_DROP_U_RUNT_MGMT)			\
+	R(RX_DROP_U_EXPECTED_MGMT)		\
+	R(RX_DROP_U_NONBCAST_BEACON)		\
+	/* 0x50 */				\
+	R(RX_DROP_U_MALFORMED_ACTION)		\
+	R(RX_DROP_U_UNKNOWN_MCAST_ACTION)	\
+	R(RX_DROP_U_UNEXPECTED_EXT_FRAME)	\
+	R(RX_DROP_U_UNHANDLED_MGMT)		\
+	R(RX_DROP_U_MCAST_DEAUTH)		\
+	R(RX_DROP_U_UNHANDLED_DEAUTH)		\
+	R(RX_DROP_U_MCAST_DISASSOC)		\
+	R(RX_DROP_U_UNHANDLED_DISASSOC)		\
+	R(RX_DROP_U_UNHANDLED_PREQ)		\
+	R(RX_DROP_U_UNHANDLED_MGMT_STYPE)	\
+	R(RX_DROP_U_NO_LINK)			\
+/* this line for the trailing \ - add before this */
+
+/* having two enums allows for checking ieee80211_rx_result use with sparse */
+enum ___mac80211_drop_reason {
+/* if we get to the end of handlers with RX_CONTINUE this will be the reason */
+	___RX_CONTINUE	= SKB_CONSUMED,
+
+/* this never gets used as an argument to kfree_skb_reason() */
+	___RX_QUEUED	= SKB_NOT_DROPPED_YET,
+
+#define ENUM(x) ___ ## x,
+	___RX_DROP_UNUSABLE = SKB_DROP_REASON_SUBSYS_MAC80211_UNUSABLE <<
+		SKB_DROP_REASON_SUBSYS_SHIFT,
+	MAC80211_DROP_REASONS_UNUSABLE(ENUM)
+#undef ENUM
+};
+
+enum mac80211_drop_reason {
+	RX_CONTINUE	= (__force ieee80211_rx_result)___RX_CONTINUE,
+	RX_QUEUED	= (__force ieee80211_rx_result)___RX_QUEUED,
+#define DEF(x) x = (__force ieee80211_rx_result)___ ## x,
+	MAC80211_DROP_REASONS_UNUSABLE(DEF)
+#undef DEF
+};
+
+#define RX_RES_IS_UNUSABLE(result)	\
+	(((__force u32)(result) & SKB_DROP_REASON_SUBSYS_MASK) == ___RX_DROP_UNUSABLE)
+
+#endif /* MAC80211_DROP_H */