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/arch/arm/boot/Makefile b/arch/arm/boot/Makefile
new file mode 100644
index 000000000..ba9b9a802
--- /dev/null
+++ b/arch/arm/boot/Makefile
@@ -0,0 +1,92 @@
+#
+# arch/arm/boot/Makefile
+#
+# This file is included by the global makefile so that you can add your own
+# architecture-specific flags and dependencies.
+#
+# This file is subject to the terms and conditions of the GNU General Public
+# License.  See the file "COPYING" in the main directory of this archive
+# for more details.
+#
+# Copyright (C) 1995-2002 Russell King
+#
+OBJCOPYFLAGS	:=-O binary -R .comment -S
+
+#   ZRELADDR == virt_to_phys(PAGE_OFFSET + TEXT_OFFSET)
+ifdef CONFIG_PHYS_OFFSET
+add_hex = $(shell printf 0x%x $$(( $(1) + $(2) )) )
+ZRELADDR    := $(call add_hex, $(CONFIG_PHYS_OFFSET), $(TEXT_OFFSET))
+endif
+
+PHYS_OFFSET := $(CONFIG_PHYS_OFFSET)
+export ZRELADDR PARAMS_PHYS PHYS_OFFSET
+
+targets := Image zImage xipImage bootpImage uImage
+
+ifeq ($(CONFIG_XIP_KERNEL),y)
+
+cmd_deflate_xip_data = $(CONFIG_SHELL) -c '$(src)/deflate_xip_data.sh $< $@'
+
+ifeq ($(CONFIG_XIP_DEFLATED_DATA),y)
+quiet_cmd_mkxip = XIPZ    $@
+cmd_mkxip = $(cmd_objcopy) && $(cmd_deflate_xip_data)
+else
+quiet_cmd_mkxip = $(quiet_cmd_objcopy)
+cmd_mkxip = $(cmd_objcopy)
+endif
+
+$(obj)/xipImage: vmlinux FORCE
+	$(call if_changed,mkxip)
+	@$(kecho) '  Physical Address of xipImage: $(CONFIG_XIP_PHYS_ADDR)'
+
+$(obj)/Image $(obj)/zImage: FORCE
+	@echo 'Kernel configured for XIP (CONFIG_XIP_KERNEL=y)'
+	@echo 'Only the xipImage target is available in this case'
+	@false
+
+else
+
+$(obj)/xipImage: FORCE
+	@echo 'Kernel not configured for XIP (CONFIG_XIP_KERNEL!=y)'
+	@false
+
+$(obj)/Image: vmlinux FORCE
+	$(call if_changed,objcopy)
+
+$(obj)/compressed/vmlinux: $(obj)/Image FORCE
+	$(Q)$(MAKE) $(build)=$(obj)/compressed $@
+
+$(obj)/zImage:	$(obj)/compressed/vmlinux FORCE
+	$(call if_changed,objcopy)
+
+endif
+
+ifneq ($(LOADADDR),)
+  UIMAGE_LOADADDR=$(LOADADDR)
+else
+  ifeq ($(CONFIG_ZBOOT_ROM),y)
+    UIMAGE_LOADADDR=$(CONFIG_ZBOOT_ROM_TEXT)
+  else
+    UIMAGE_LOADADDR=$(ZRELADDR)
+  endif
+endif
+
+check_for_multiple_loadaddr = \
+if [ $(words $(UIMAGE_LOADADDR)) -ne 1 ]; then \
+	echo 'multiple (or no) load addresses: $(UIMAGE_LOADADDR)'; \
+	echo 'This is incompatible with uImages'; \
+	echo 'Specify LOADADDR on the commandline to build an uImage'; \
+	false; \
+fi
+
+$(obj)/uImage:	$(obj)/zImage FORCE
+	@$(check_for_multiple_loadaddr)
+	$(call if_changed,uimage)
+
+$(obj)/bootp/bootp: $(obj)/zImage FORCE
+	$(Q)$(MAKE) $(build)=$(obj)/bootp $@
+
+$(obj)/bootpImage: $(obj)/bootp/bootp FORCE
+	$(call if_changed,objcopy)
+
+subdir-	    := bootp compressed dts