sched_ext: Maintain an online cid mask in the scheduler arenagrafted

Schedulers on the default cid mapping treat [0, nr_online_cids) as the
online set and restart on hotplug. Schedulers that install their own mapping
with scx_bpf_cid_override() have no way to learn which cids are online: the
count no longer identifies members and the CPU-form cpumask is unusable from
cid programs. This is an obvious hole in the cid API.

Add scx_bpf_online_cmask(), a kernel-maintained cmask in the scheduler's
arena, allocated alongside the per-CPU scratch masks and populated after the
cid mapping is finalized and before ops.init(), for child schedulers too.
The pointer stays valid through ops.exit() with no reference to take. It is
the arena offset as a void pointer, the same form struct_ops arena arguments
arrive in. The verifier types the void return as a scalar for the program's
arena cast.

The mask follows the SCX hotplug notifications: seeded from cpu_active_mask
and updated before ops.cid_online/offline() runs, so it lags cpu_online_mask
only inside a hotplug transition. Updates walk the scheduler list under the
lock that also serializes unlinking. Reads are live, not atomic snapshots.
Root initialization excludes hotplug.

v2: Reworded the getter kerneldoc (Andrea Righi).

Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>

1 files changed, 143 insertions, 0 deletions

diff --git a/lib/tests/liveupdate.c b/lib/tests/liveupdate.c
new file mode 100644
index 000000000..4c08a7c6f
--- /dev/null
+++ b/lib/tests/liveupdate.c
@@ -0,0 +1,143 @@
+// SPDX-License-Identifier: GPL-2.0
+
+/*
+ * Copyright (c) 2025, Google LLC.
+ * Pasha Tatashin <pasha.tatashin@soleen.com>
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME " test: " fmt
+
+#include <linux/cleanup.h>
+#include <linux/errno.h>
+#include <linux/init.h>
+#include <linux/liveupdate.h>
+#include <linux/module.h>
+#include "../../kernel/liveupdate/luo_internal.h"
+
+static const struct liveupdate_flb_ops test_flb_ops;
+#define DEFINE_TEST_FLB(i) {						\
+	.ops = &test_flb_ops,						\
+	.compatible = LIVEUPDATE_TEST_FLB_COMPATIBLE(i),		\
+}
+
+/* Number of Test FLBs to register with every file handler */
+#define TEST_NFLBS 3
+static struct liveupdate_flb test_flbs[TEST_NFLBS] = {
+	DEFINE_TEST_FLB(0),
+	DEFINE_TEST_FLB(1),
+	DEFINE_TEST_FLB(2),
+};
+
+#define TEST_FLB_MAGIC_BASE 0xFEEDF00DCAFEBEE0ULL
+
+static int test_flb_preserve(struct liveupdate_flb_op_args *argp)
+{
+	ptrdiff_t index = argp->flb - test_flbs;
+
+	pr_info("%s: preserve was triggered\n", argp->flb->compatible);
+	argp->data = TEST_FLB_MAGIC_BASE + index;
+
+	return 0;
+}
+
+static void test_flb_unpreserve(struct liveupdate_flb_op_args *argp)
+{
+	pr_info("%s: unpreserve was triggered\n", argp->flb->compatible);
+}
+
+static int test_flb_retrieve(struct liveupdate_flb_op_args *argp)
+{
+	ptrdiff_t index = argp->flb - test_flbs;
+	u64 expected_data = TEST_FLB_MAGIC_BASE + index;
+
+	if (argp->data == expected_data) {
+		pr_info("%s: found flb data from the previous boot\n",
+			argp->flb->compatible);
+		argp->obj = (void *)argp->data;
+	} else {
+		pr_err("%s: ERROR - incorrect data handle: %llx, expected %llx\n",
+		       argp->flb->compatible, argp->data, expected_data);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+static void test_flb_finish(struct liveupdate_flb_op_args *argp)
+{
+	ptrdiff_t index = argp->flb - test_flbs;
+	void *expected_obj = (void *)(TEST_FLB_MAGIC_BASE + index);
+
+	if (argp->obj == expected_obj) {
+		pr_info("%s: finish was triggered\n", argp->flb->compatible);
+	} else {
+		pr_err("%s: ERROR - finish called with invalid object\n",
+		       argp->flb->compatible);
+	}
+}
+
+static const struct liveupdate_flb_ops test_flb_ops = {
+	.preserve	= test_flb_preserve,
+	.unpreserve	= test_flb_unpreserve,
+	.retrieve	= test_flb_retrieve,
+	.finish		= test_flb_finish,
+	.owner		= THIS_MODULE,
+};
+
+static void liveupdate_test_init(void)
+{
+	static DEFINE_MUTEX(init_lock);
+	static bool initialized;
+	int i;
+
+	guard(mutex)(&init_lock);
+
+	if (initialized)
+		return;
+
+	for (i = 0; i < TEST_NFLBS; i++) {
+		struct liveupdate_flb *flb = &test_flbs[i];
+		void *obj;
+		int err;
+
+		err = liveupdate_flb_get_incoming(flb, &obj);
+		if (err && err != -ENODATA && err != -ENOENT) {
+			pr_err("liveupdate_flb_get_incoming for %s failed: %pe\n",
+			       flb->compatible, ERR_PTR(err));
+		}
+
+		if (!err)
+			liveupdate_flb_put_incoming(flb);
+	}
+	initialized = true;
+}
+
+void liveupdate_test_register(struct liveupdate_file_handler *fh)
+{
+	int err, i;
+
+	liveupdate_test_init();
+
+	for (i = 0; i < TEST_NFLBS; i++) {
+		struct liveupdate_flb *flb = &test_flbs[i];
+
+		err = liveupdate_register_flb(fh, flb);
+		if (err) {
+			pr_err("Failed to register %s %pe\n",
+			       flb->compatible, ERR_PTR(err));
+		}
+	}
+
+	err = liveupdate_register_flb(fh, &test_flbs[0]);
+	if (!err || err != -EEXIST) {
+		pr_err("Failed: %s should be already registered, but got err: %pe\n",
+		       test_flbs[0].compatible, ERR_PTR(err));
+	}
+
+	pr_info("Registered %d FLBs with file handler: [%s]\n",
+		TEST_NFLBS, fh->compatible);
+}
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Pasha Tatashin <pasha.tatashin@soleen.com>");
+MODULE_DESCRIPTION("In-kernel test for LUO mechanism");