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, 25 insertions, 0 deletions

diff --git a/drivers/net/ipa/Makefile b/drivers/net/ipa/Makefile
new file mode 100644
index 000000000..e148ec3c1
--- /dev/null
+++ b/drivers/net/ipa/Makefile
@@ -0,0 +1,25 @@
+# SPDX-License-Identifier: GPL-2.0
+#
+# Makefile for the Qualcomm IPA driver.
+
+IPA_REG_VERSIONS	:=	3.1 3.5.1 4.2 4.5 4.7 4.9 4.11 5.0 5.5
+
+# Some IPA versions can reuse another set of GSI register definitions.
+GSI_REG_VERSIONS	:=	3.1 3.5.1 4.0 4.5 4.9 4.11 5.0
+
+IPA_DATA_VERSIONS	:=	3.1 3.5.1 4.2 4.5 4.7 4.9 4.11 5.0 5.2 5.5
+
+obj-$(CONFIG_QCOM_IPA)	+=	ipa.o
+
+ipa-y			:=	ipa_main.o ipa_power.o ipa_reg.o ipa_mem.o \
+				ipa_table.o ipa_interrupt.o gsi.o gsi_reg.o \
+				gsi_trans.o ipa_gsi.o ipa_smp2p.o ipa_uc.o \
+				ipa_endpoint.o ipa_cmd.o ipa_modem.o \
+				ipa_resource.o ipa_qmi.o ipa_qmi_msg.o \
+				ipa_sysfs.o
+
+ipa-y			+=	$(IPA_REG_VERSIONS:%=reg/ipa_reg-v%.o)
+
+ipa-y			+=	$(GSI_REG_VERSIONS:%=reg/gsi_reg-v%.o)
+
+ipa-y			+=	$(IPA_DATA_VERSIONS:%=data/ipa_data-v%.o)