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, 168 insertions, 0 deletions

diff --git a/drivers/edac/edac_module.c b/drivers/edac/edac_module.c
new file mode 100644
index 000000000..1c9f62382
--- /dev/null
+++ b/drivers/edac/edac_module.c
@@ -0,0 +1,168 @@
+/*
+ * edac_module.c
+ *
+ * (C) 2007 www.softwarebitmaker.com
+ *
+ * This file is licensed under the terms of the GNU General Public
+ * License version 2. This program is licensed "as is" without any
+ * warranty of any kind, whether express or implied.
+ *
+ * Author: Doug Thompson <dougthompson@xmission.com>
+ *
+ */
+#include <linux/edac.h>
+
+#include "edac_mc.h"
+#include "edac_module.h"
+
+#define EDAC_VERSION "Ver: 3.0.0"
+
+#ifdef CONFIG_EDAC_DEBUG
+
+static int edac_set_debug_level(const char *buf,
+				const struct kernel_param *kp)
+{
+	unsigned long val;
+	int ret;
+
+	ret = kstrtoul(buf, 0, &val);
+	if (ret)
+		return ret;
+
+	if (val > 4)
+		return -EINVAL;
+
+	return param_set_int(buf, kp);
+}
+
+/* Values of 0 to 4 will generate output */
+int edac_debug_level = 2;
+EXPORT_SYMBOL_GPL(edac_debug_level);
+
+module_param_call(edac_debug_level, edac_set_debug_level, param_get_int,
+		  &edac_debug_level, 0644);
+MODULE_PARM_DESC(edac_debug_level, "EDAC debug level: [0-4], default: 2");
+#endif
+
+/*
+ * edac_op_state_to_string()
+ */
+char *edac_op_state_to_string(int opstate)
+{
+	if (opstate == OP_RUNNING_POLL)
+		return "POLLED";
+	else if (opstate == OP_RUNNING_INTERRUPT)
+		return "INTERRUPT";
+	else if (opstate == OP_RUNNING_POLL_INTR)
+		return "POLL-INTR";
+	else if (opstate == OP_ALLOC)
+		return "ALLOC";
+	else if (opstate == OP_OFFLINE)
+		return "OFFLINE";
+
+	return "UNKNOWN";
+}
+
+/*
+ * sysfs object: /sys/devices/system/edac
+ *	need to export to other files
+ */
+static const struct bus_type edac_subsys = {
+	.name = "edac",
+	.dev_name = "edac",
+};
+
+static int edac_subsys_init(void)
+{
+	int err;
+
+	/* create the /sys/devices/system/edac directory */
+	err = subsys_system_register(&edac_subsys, NULL);
+	if (err)
+		printk(KERN_ERR "Error registering toplevel EDAC sysfs dir\n");
+
+	return err;
+}
+
+static void edac_subsys_exit(void)
+{
+	bus_unregister(&edac_subsys);
+}
+
+/* return pointer to the 'edac' node in sysfs */
+const struct bus_type *edac_get_sysfs_subsys(void)
+{
+	return &edac_subsys;
+}
+EXPORT_SYMBOL_GPL(edac_get_sysfs_subsys);
+/*
+ * edac_init
+ *      module initialization entry point
+ */
+static int __init edac_init(void)
+{
+	int err = 0;
+
+	edac_printk(KERN_INFO, EDAC_MC, EDAC_VERSION "\n");
+
+	err = edac_subsys_init();
+	if (err)
+		return err;
+
+	/*
+	 * Harvest and clear any boot/initialization PCI parity errors
+	 *
+	 * FIXME: This only clears errors logged by devices present at time of
+	 *      module initialization.  We should also do an initial clear
+	 *      of each newly hotplugged device.
+	 */
+	edac_pci_clear_parity_errors();
+
+	err = edac_mc_sysfs_init();
+	if (err)
+		goto err_sysfs;
+
+	edac_debugfs_init();
+
+	err = edac_workqueue_setup();
+	if (err) {
+		edac_printk(KERN_ERR, EDAC_MC, "Failure initializing workqueue\n");
+		goto err_wq;
+	}
+
+	return 0;
+
+err_wq:
+	edac_debugfs_exit();
+	edac_mc_sysfs_exit();
+
+err_sysfs:
+	edac_subsys_exit();
+
+	return err;
+}
+
+/*
+ * edac_exit()
+ *      module exit/termination function
+ */
+static void __exit edac_exit(void)
+{
+	edac_dbg(0, "\n");
+
+	/* tear down the various subsystems */
+	edac_workqueue_teardown();
+	edac_mc_sysfs_exit();
+	edac_debugfs_exit();
+	edac_subsys_exit();
+}
+
+/*
+ * Inform the kernel of our entry and exit points
+ */
+subsys_initcall(edac_init);
+module_exit(edac_exit);
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Doug Thompson www.softwarebitmaker.com, et al");
+MODULE_DESCRIPTION("Core library routines for EDAC reporting");