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, 48 insertions, 0 deletions

diff --git a/drivers/misc/eeprom/Kconfig b/drivers/misc/eeprom/Kconfig
new file mode 100644
index 000000000..d6af02588
--- /dev/null
+++ b/drivers/misc/eeprom/Kconfig
@@ -0,0 +1,48 @@
+# SPDX-License-Identifier: GPL-2.0-only
+menu "EEPROM support"
+
+config EEPROM_MAX6875
+	tristate "Maxim MAX6874/5 power supply supervisor"
+	depends on I2C
+	help
+	  If you say yes here you get read-only support for the user EEPROM of
+	  the Maxim MAX6874/5 EEPROM-programmable, quad power-supply
+	  sequencer/supervisor.
+
+	  All other features of this chip should be accessed via i2c-dev.
+
+	  This driver can also be built as a module.  If so, the module
+	  will be called max6875.
+
+
+config EEPROM_93CX6
+	tristate "EEPROM 93CX6 support"
+	help
+	  This is a driver for the EEPROM chipsets 93c46 and 93c66.
+	  The driver supports both read as well as write commands.
+
+	  If unsure, say N.
+
+config EEPROM_DIGSY_MTC_CFG
+	bool "DigsyMTC display configuration EEPROMs device"
+	depends on GPIO_MPC5200 && SPI_GPIO
+	help
+	  This option enables access to display configuration EEPROMs
+	  on digsy_mtc board. You have to additionally select Microwire
+	  EEPROM 93XX46 driver. sysfs entries will be created for that
+	  EEPROM allowing to read/write the configuration data or to
+	  erase the whole EEPROM.
+
+	  If unsure, say N.
+
+config EEPROM_IDT_89HPESX
+	tristate "IDT 89HPESx PCIe-switches EEPROM / CSR support"
+	depends on I2C && SYSFS
+	help
+	  Enable this driver to get read/write access to EEPROM / CSRs
+	  over IDT PCIe-switch i2c-slave interface.
+
+	  This driver can also be built as a module. If so, the module
+	  will be called idt_89hpesx.
+
+endmenu