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, 85 insertions, 0 deletions

diff --git a/usr/Makefile b/usr/Makefile
new file mode 100644
index 000000000..e8f42478a
--- /dev/null
+++ b/usr/Makefile
@@ -0,0 +1,85 @@
+# SPDX-License-Identifier: GPL-2.0
+#
+# kbuild file for usr/ - including initramfs image
+#
+
+compress-y					:= copy
+compress-$(CONFIG_INITRAMFS_COMPRESSION_GZIP)	:= gzip
+compress-$(CONFIG_INITRAMFS_COMPRESSION_BZIP2)	:= bzip2
+compress-$(CONFIG_INITRAMFS_COMPRESSION_LZMA)	:= lzma
+compress-$(CONFIG_INITRAMFS_COMPRESSION_XZ)	:= xzmisc
+compress-$(CONFIG_INITRAMFS_COMPRESSION_LZO)	:= lzo
+compress-$(CONFIG_INITRAMFS_COMPRESSION_LZ4)	:= lz4
+compress-$(CONFIG_INITRAMFS_COMPRESSION_ZSTD)	:= zstd
+
+obj-$(CONFIG_BLK_DEV_INITRD) := initramfs_data.o
+
+$(obj)/initramfs_data.o: $(obj)/initramfs_inc_data
+
+hostprogs := gen_init_cpio
+
+ramfs-input := $(CONFIG_INITRAMFS_SOURCE)
+cpio-data :=
+
+# If CONFIG_INITRAMFS_SOURCE is empty, generate a small initramfs with the
+# default contents.
+ifeq ($(ramfs-input),)
+ramfs-input := $(src)/default_cpio_list
+endif
+
+ifeq ($(words $(ramfs-input)),1)
+
+# If CONFIG_INITRAMFS_SOURCE specifies a single file, and it is suffixed with
+# .cpio, use it directly as an initramfs.
+ifneq ($(filter %.cpio,$(ramfs-input)),)
+cpio-data := $(ramfs-input)
+endif
+
+# If CONFIG_INITRAMFS_SOURCE specifies a single file, and it is suffixed with
+# .cpio.*, use it directly as an initramfs, and avoid double compression.
+ifeq ($(words $(subst .cpio.,$(space),$(ramfs-input))),2)
+cpio-data := $(ramfs-input)
+compress-y := copy
+endif
+
+endif
+
+# For other cases, generate the initramfs cpio archive based on the contents
+# specified by CONFIG_INITRAMFS_SOURCE.
+ifeq ($(cpio-data),)
+
+cpio-data := $(obj)/initramfs_data.cpio
+
+# .initramfs_data.cpio.d is used to identify all files included
+# in initramfs and to detect if any files are added/removed.
+# Removed files are identified by directory timestamp being updated
+# The dependency list is generated by gen_initramfs.sh -l
+-include $(obj)/.initramfs_data.cpio.d
+
+# do not try to update files included in initramfs
+$(deps_initramfs): ;
+
+quiet_cmd_initfs = GEN     $@
+      cmd_initfs = \
+	$(CONFIG_SHELL) $< -o $@ -l $(obj)/.initramfs_data.cpio.d \
+	$(addprefix -u , $(CONFIG_INITRAMFS_ROOT_UID)) \
+	$(addprefix -g , $(CONFIG_INITRAMFS_ROOT_GID)) \
+	$(if $(KBUILD_BUILD_TIMESTAMP), -d "$(KBUILD_BUILD_TIMESTAMP)") \
+	$(ramfs-input)
+
+# We rebuild initramfs_data.cpio if:
+# 1) Any included file is newer than initramfs_data.cpio
+# 2) There are changes in which files are included (added or deleted)
+# 3) If gen_init_cpio are newer than initramfs_data.cpio
+# 4) Arguments to gen_initramfs.sh changes
+$(obj)/initramfs_data.cpio: $(src)/gen_initramfs.sh $(obj)/gen_init_cpio $(deps_initramfs) FORCE
+	$(call if_changed,initfs)
+
+endif
+
+$(obj)/initramfs_inc_data: $(cpio-data) FORCE
+	$(call if_changed,$(compress-y))
+
+targets += initramfs_data.cpio initramfs_inc_data
+
+subdir-$(CONFIG_UAPI_HEADER_TEST) += include