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, 42 insertions, 0 deletions

diff --git a/scripts/Makefile.clang b/scripts/Makefile.clang
new file mode 100644
index 000000000..b67636b28
--- /dev/null
+++ b/scripts/Makefile.clang
@@ -0,0 +1,42 @@
+# Individual arch/{arch}/Makefiles should use -EL/-EB to set intended
+# endianness and -m32/-m64 to set word size based on Kconfigs instead of
+# relying on the target triple.
+CLANG_TARGET_FLAGS_arm		:= arm-linux-gnueabi
+CLANG_TARGET_FLAGS_arm64	:= aarch64-linux-gnu
+CLANG_TARGET_FLAGS_hexagon	:= hexagon-linux-musl
+CLANG_TARGET_FLAGS_loongarch	:= loongarch64-linux-gnusf
+CLANG_TARGET_FLAGS_m68k		:= m68k-linux-gnu
+CLANG_TARGET_FLAGS_mips		:= mipsel-linux-gnu
+CLANG_TARGET_FLAGS_powerpc	:= powerpc64le-linux-gnu
+CLANG_TARGET_FLAGS_riscv	:= riscv64-linux-gnu
+CLANG_TARGET_FLAGS_s390		:= s390x-linux-gnu
+CLANG_TARGET_FLAGS_sparc	:= sparc64-linux-gnu
+CLANG_TARGET_FLAGS_x86		:= x86_64-linux-gnu
+# This is only for i386 UM builds, which need the 32-bit target not -m32
+CLANG_TARGET_FLAGS_i386		:= i386-linux-gnu
+CLANG_TARGET_FLAGS_um		:= $(CLANG_TARGET_FLAGS_$(SUBARCH))
+CLANG_TARGET_FLAGS		:= $(CLANG_TARGET_FLAGS_$(SRCARCH))
+
+ifeq ($(CLANG_TARGET_FLAGS),)
+$(error add '--target=' option to scripts/Makefile.clang)
+else
+CLANG_FLAGS	+= --target=$(CLANG_TARGET_FLAGS)
+endif
+
+ifeq ($(LLVM_IAS),0)
+CLANG_FLAGS	+= -fno-integrated-as
+GCC_TOOLCHAIN_DIR := $(dir $(shell which $(CROSS_COMPILE)elfedit))
+CLANG_FLAGS	+= --prefix=$(GCC_TOOLCHAIN_DIR)$(notdir $(CROSS_COMPILE))
+else
+CLANG_FLAGS	+= -fintegrated-as
+endif
+# By default, clang only warns when it encounters an unknown warning flag or
+# certain optimization flags it knows it has not implemented.
+# Make it behave more like gcc by erroring when these flags are encountered
+# so they can be implemented or wrapped in cc-option.
+CLANG_FLAGS	+= -Werror=unknown-warning-option
+CLANG_FLAGS	+= -Werror=ignored-optimization-argument
+CLANG_FLAGS	+= -Werror=option-ignored
+CLANG_FLAGS	+= -Werror=unused-command-line-argument
+KBUILD_CPPFLAGS	+= $(CLANG_FLAGS)
+export CLANG_FLAGS