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, 98 insertions, 0 deletions

diff --git a/scripts/syscallhdr.sh b/scripts/syscallhdr.sh
new file mode 100755
index 000000000..22e34cd46
--- /dev/null
+++ b/scripts/syscallhdr.sh
@@ -0,0 +1,98 @@
+#!/bin/sh
+# SPDX-License-Identifier: GPL-2.0-only
+#
+# Generate a syscall number header.
+#
+# Each line of the syscall table should have the following format:
+#
+# NR ABI NAME [NATIVE] [COMPAT]
+#
+# NR       syscall number
+# ABI      ABI name
+# NAME     syscall name
+# NATIVE   native entry point (optional)
+# COMPAT   compat entry point (optional)
+
+set -e
+
+usage() {
+	echo >&2 "usage: $0 [--abis ABIS] [--emit-nr] [--offset OFFSET] [--prefix PREFIX] INFILE OUTFILE" >&2
+	echo >&2
+	echo >&2 "  INFILE    input syscall table"
+	echo >&2 "  OUTFILE   output header file"
+	echo >&2
+	echo >&2 "options:"
+	echo >&2 "  --abis ABIS        ABI(s) to handle (By default, all lines are handled)"
+	echo >&2 "  --emit-nr          Emit the macro of the number of syscalls (__NR_syscalls)"
+	echo >&2 "  --offset OFFSET    The offset of syscall numbers"
+	echo >&2 "  --prefix PREFIX    The prefix to the macro like __NR_<PREFIX><NAME>"
+	exit 1
+}
+
+# default unless specified by options
+abis=
+emit_nr=
+offset=
+prefix=
+
+while [ $# -gt 0 ]
+do
+	case $1 in
+	--abis)
+		abis=$(echo "($2)" | tr ',' '|')
+		shift 2;;
+	--emit-nr)
+		emit_nr=1
+		shift 1;;
+	--offset)
+		offset=$2
+		shift 2;;
+	--prefix)
+		prefix=$2
+		shift 2;;
+	-*)
+		echo "$1: unknown option" >&2
+		usage;;
+	*)
+		break;;
+	esac
+done
+
+if [ $# -ne 2 ]; then
+	usage
+fi
+
+infile="$1"
+outfile="$2"
+
+guard=_UAPI_ASM_$(basename "$outfile" |
+	sed -e 'y/abcdefghijklmnopqrstuvwxyz/ABCDEFGHIJKLMNOPQRSTUVWXYZ/' \
+	-e 's/[^A-Z0-9_]/_/g' -e 's/__/_/g')
+
+grep -E "^[0-9A-Fa-fXx]+[[:space:]]+$abis" "$infile" | {
+	echo "#ifndef $guard"
+	echo "#define $guard"
+	echo
+
+	max=0
+	while read nr abi name native compat ; do
+
+		max=$nr
+
+		if [ -n "$offset" ]; then
+			nr="($offset + $nr)"
+		fi
+
+		echo "#define __NR_$prefix$name $nr"
+	done
+
+	if [ -n "$emit_nr" ]; then
+		echo
+		echo "#ifdef __KERNEL__"
+		echo "#define __NR_${prefix}syscalls $(($max + 1))"
+		echo "#endif"
+	fi
+
+	echo
+	echo "#endif /* $guard */"
+} > "$outfile"