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, 90 insertions, 0 deletions

diff --git a/tools/perf/ui/util.c b/tools/perf/ui/util.c
new file mode 100644
index 000000000..1d38ddf01
--- /dev/null
+++ b/tools/perf/ui/util.c
@@ -0,0 +1,90 @@
+// SPDX-License-Identifier: GPL-2.0
+#include "util.h"
+#include "../util/debug.h"
+#include <stdio.h>
+
+/*
+ * Default error logging functions
+ */
+static int perf_stdio__error(const char *format, va_list args)
+{
+	fprintf(stderr, "Error:\n");
+	vfprintf(stderr, format, args);
+	return 0;
+}
+
+static int perf_stdio__warning(const char *format, va_list args)
+{
+	if (quiet)
+		return 0;
+
+	fprintf(stderr, "Warning:\n");
+	vfprintf(stderr, format, args);
+	return 0;
+}
+
+static struct perf_error_ops default_eops =
+{
+	.error		= perf_stdio__error,
+	.warning	= perf_stdio__warning,
+};
+
+static struct perf_error_ops *perf_eops = &default_eops;
+
+
+int ui__error(const char *format, ...)
+{
+	int ret;
+	va_list args;
+
+	va_start(args, format);
+	ret = perf_eops->error(format, args);
+	va_end(args);
+
+	return ret;
+}
+
+int ui__warning(const char *format, ...)
+{
+	int ret;
+	va_list args;
+	if (quiet)
+		return 0;
+
+	va_start(args, format);
+	ret = perf_eops->warning(format, args);
+	va_end(args);
+
+	return ret;
+}
+
+/**
+ * perf_error__register - Register error logging functions
+ * @eops: The pointer to error logging function struct
+ *
+ * Register UI-specific error logging functions. Before calling this,
+ * other logging functions should be unregistered, if any.
+ */
+int perf_error__register(struct perf_error_ops *eops)
+{
+	if (perf_eops != &default_eops)
+		return -1;
+
+	perf_eops = eops;
+	return 0;
+}
+
+/**
+ * perf_error__unregister - Unregister error logging functions
+ * @eops: The pointer to error logging function struct
+ *
+ * Unregister already registered error logging functions.
+ */
+int perf_error__unregister(struct perf_error_ops *eops)
+{
+	if (perf_eops != eops)
+		return -1;
+
+	perf_eops = &default_eops;
+	return 0;
+}