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, 83 insertions, 0 deletions

diff --git a/drivers/net/pse-pd/Kconfig b/drivers/net/pse-pd/Kconfig
new file mode 100644
index 000000000..a0f2ae668
--- /dev/null
+++ b/drivers/net/pse-pd/Kconfig
@@ -0,0 +1,83 @@
+# SPDX-License-Identifier: GPL-2.0-only
+#
+# Ethernet Power Sourcing Equipment drivers
+#
+
+menuconfig PSE_CONTROLLER
+	bool "Ethernet Power Sourcing Equipment Support"
+	depends on REGULATOR
+	help
+	  Generic Power Sourcing Equipment Controller support.
+
+	  If unsure, say no.
+
+if PSE_CONTROLLER
+
+config PSE_REALTEK_MCU
+	tristate
+	help
+	  Shared core for the Realtek PSE MCU driver. This is selected
+	  automatically by the transport options below.
+
+config PSE_REALTEK_MCU_I2C
+	tristate "Realtek PSE MCU driver (I2C transport)"
+	depends on I2C
+	select PSE_REALTEK_MCU
+	help
+	  Driver for the microcontroller (MCU) that fronts the PSE
+	  hardware on various Realtek-based managed switches, attached
+	  via I2C/SMBus. The MCU exposes a message-based protocol; the actual
+	  PSE silicon is not accessed directly. To compile this driver as a
+	  module, choose M here: the module will be called realtek-pse-mcu-i2c.
+
+config PSE_REALTEK_MCU_UART
+	tristate "Realtek PSE MCU driver (UART transport)"
+	depends on SERIAL_DEV_BUS
+	select PSE_REALTEK_MCU
+	help
+	  Driver for the microcontroller (MCU) that fronts the PSE
+	  hardware on various Realtek-based managed switches, attached
+	  via UART. The MCU exposes a message-based protocol; the actual PSE
+	  silicon is not accessed directly. To compile this driver as a
+	  module, choose M here: the module will be called realtek-pse-mcu-uart.
+
+config PSE_REGULATOR
+	tristate "Regulator based PSE controller"
+	help
+	  This module provides support for simple regulator based Ethernet Power
+	  Sourcing Equipment without automatic classification support. For
+	  example for basic implementation of PoDL (802.3bu) specification.
+
+config PSE_PD692X0
+	tristate "PD692X0 PSE controller"
+	depends on I2C
+	select FW_LOADER
+	select FW_UPLOAD
+	help
+	  This module provides support for PD692x0 regulator based Ethernet
+	  Power Sourcing Equipment.
+
+	  To compile this driver as a module, choose M here: the
+	  module will be called pd692x0.
+
+config PSE_SI3474
+	tristate "Si3474 PSE controller"
+	depends on I2C
+	help
+	  This module provides support for Si3474 regulator based Ethernet
+	  Power Sourcing Equipment.
+	  Only 4-pair PSE configurations are supported.
+
+	  To compile this driver as a module, choose M here: the
+	  module will be called si3474.
+
+config PSE_TPS23881
+	tristate "TPS23881 PSE controller"
+	depends on I2C
+	help
+	  This module provides support for TPS23881 regulator based Ethernet
+	  Power Sourcing Equipment.
+
+	  To compile this driver as a module, choose M here: the
+	  module will be called tps23881.
+endif