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, 40 insertions, 0 deletions

diff --git a/arch/hexagon/Makefile b/arch/hexagon/Makefile
new file mode 100644
index 000000000..ff172cbe5
--- /dev/null
+++ b/arch/hexagon/Makefile
@@ -0,0 +1,40 @@
+# SPDX-License-Identifier: GPL-2.0
+#  Makefile for the Hexagon arch
+
+KBUILD_DEFCONFIG = comet_defconfig
+
+# Do not use GP-relative jumps
+KBUILD_CFLAGS += -G0
+LDFLAGS_vmlinux += -G0
+
+# Do not use single-byte enums; these will overflow.
+KBUILD_CFLAGS += -fno-short-enums
+
+# We must use long-calls:
+KBUILD_CFLAGS += -mlong-calls
+
+# Modules must use either long-calls, or use pic/plt.
+# Use long-calls for now, it's easier.  And faster.
+# KBUILD_CFLAGS_MODULE += -fPIC
+# KBUILD_LDFLAGS_MODULE += -shared
+KBUILD_CFLAGS_MODULE += -mlong-calls
+
+cflags-y += $(call cc-option,-mv${CONFIG_HEXAGON_ARCH_VERSION})
+aflags-y += $(call cc-option,-mv${CONFIG_HEXAGON_ARCH_VERSION})
+ldflags-y += $(call cc-option,-mv${CONFIG_HEXAGON_ARCH_VERSION})
+
+KBUILD_CFLAGS += $(cflags-y)
+KBUILD_AFLAGS += $(aflags-y)
+KBUILD_LDFLAGS += $(ldflags-y)
+
+# Thread-info register will be r19.  This value is not configureable;
+# it is hard-coded in several files.
+TIR_NAME := r19
+KBUILD_CFLAGS += -ffixed-$(TIR_NAME) -DTHREADINFO_REG=$(TIR_NAME) -D__linux__
+KBUILD_AFLAGS += -DTHREADINFO_REG=$(TIR_NAME)
+
+# Disable HexagonConstExtenders pass for LLVM versions prior to 19.1.0
+# https://github.com/llvm/llvm-project/issues/99714
+ifneq ($(call clang-min-version, 190100),y)
+KBUILD_CFLAGS += -mllvm -hexagon-cext=false
+endif