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, 134 insertions, 0 deletions

diff --git a/include/trace/events/module.h b/include/trace/events/module.h
new file mode 100644
index 000000000..e5a006be9
--- /dev/null
+++ b/include/trace/events/module.h
@@ -0,0 +1,134 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Because linux/module.h has tracepoints in the header, and ftrace.h
+ * used to include this file, define_trace.h includes linux/module.h
+ * But we do not want the module.h to override the TRACE_SYSTEM macro
+ * variable that define_trace.h is processing, so we only set it
+ * when module events are being processed, which would happen when
+ * CREATE_TRACE_POINTS is defined.
+ */
+#ifdef CREATE_TRACE_POINTS
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM module
+#endif
+
+#if !defined(_TRACE_MODULE_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_MODULE_H
+
+#include <linux/tracepoint.h>
+
+#ifdef CONFIG_MODULES
+
+struct module;
+
+#define show_module_flags(flags) __print_flags(flags, "",	\
+	{ (1UL << TAINT_PROPRIETARY_MODULE),	"P" },		\
+	{ (1UL << TAINT_OOT_MODULE),		"O" },		\
+	{ (1UL << TAINT_FORCED_MODULE),		"F" },		\
+	{ (1UL << TAINT_CRAP),			"C" },		\
+	{ (1UL << TAINT_UNSIGNED_MODULE),	"E" })
+
+TRACE_EVENT(module_load,
+
+	TP_PROTO(struct module *mod),
+
+	TP_ARGS(mod),
+
+	TP_STRUCT__entry(
+		__field(	unsigned int,	taints		)
+		__string(	name,		mod->name	)
+	),
+
+	TP_fast_assign(
+		__entry->taints = mod->taints;
+		__assign_str(name);
+	),
+
+	TP_printk("%s %s", __get_str(name), show_module_flags(__entry->taints))
+);
+
+TRACE_EVENT(module_free,
+
+	TP_PROTO(struct module *mod),
+
+	TP_ARGS(mod),
+
+	TP_STRUCT__entry(
+		__string(	name,		mod->name	)
+	),
+
+	TP_fast_assign(
+		__assign_str(name);
+	),
+
+	TP_printk("%s", __get_str(name))
+);
+
+#ifdef CONFIG_MODULE_UNLOAD
+/* trace_module_get/put are only used if CONFIG_MODULE_UNLOAD is defined */
+
+DECLARE_EVENT_CLASS(module_refcnt,
+
+	TP_PROTO(struct module *mod, unsigned long ip),
+
+	TP_ARGS(mod, ip),
+
+	TP_STRUCT__entry(
+		__field(	unsigned long,	ip		)
+		__field(	int,		refcnt		)
+		__string(	name,		mod->name	)
+	),
+
+	TP_fast_assign(
+		__entry->ip	= ip;
+		__entry->refcnt	= atomic_read(&mod->refcnt);
+		__assign_str(name);
+	),
+
+	TP_printk("%s call_site=%ps refcnt=%d",
+		  __get_str(name), (void *)__entry->ip, __entry->refcnt)
+);
+
+DEFINE_EVENT(module_refcnt, module_get,
+
+	TP_PROTO(struct module *mod, unsigned long ip),
+
+	TP_ARGS(mod, ip)
+);
+
+DEFINE_EVENT(module_refcnt, module_put,
+
+	TP_PROTO(struct module *mod, unsigned long ip),
+
+	TP_ARGS(mod, ip)
+);
+#endif /* CONFIG_MODULE_UNLOAD */
+
+TRACE_EVENT(module_request,
+
+	TP_PROTO(char *name, bool wait, unsigned long ip),
+
+	TP_ARGS(name, wait, ip),
+
+	TP_STRUCT__entry(
+		__field(	unsigned long,	ip		)
+		__field(	bool,		wait		)
+		__string(	name,		name		)
+	),
+
+	TP_fast_assign(
+		__entry->ip	= ip;
+		__entry->wait	= wait;
+		__assign_str(name);
+	),
+
+	TP_printk("%s wait=%d call_site=%ps",
+		  __get_str(name), (int)__entry->wait, (void *)__entry->ip)
+);
+
+#endif /* CONFIG_MODULES */
+
+#endif /* _TRACE_MODULE_H */
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>