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, 151 insertions, 0 deletions

diff --git a/tools/perf/tests/thread-map.c b/tools/perf/tests/thread-map.c
new file mode 100644
index 000000000..877868107
--- /dev/null
+++ b/tools/perf/tests/thread-map.c
@@ -0,0 +1,151 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <stdlib.h>
+#include <string.h>
+#include <sys/types.h>
+#include <unistd.h>
+#include <sys/prctl.h>
+#include "tests.h"
+#include "thread_map.h"
+#include "debug.h"
+#include "event.h"
+#include "util/synthetic-events.h"
+#include <perf/event.h>
+#include <internal/threadmap.h>
+
+struct perf_sample;
+struct perf_tool;
+struct machine;
+
+#define NAME	(const char *) "perf"
+#define NAMEUL	(unsigned long) NAME
+
+static int test__thread_map(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
+{
+	struct perf_thread_map *map;
+
+	TEST_ASSERT_VAL("failed to set process name",
+			!prctl(PR_SET_NAME, NAMEUL, 0, 0, 0));
+
+	/* test map on current pid */
+	map = thread_map__new_by_pid(getpid());
+	TEST_ASSERT_VAL("failed to alloc map", map);
+
+	thread_map__read_comms(map);
+
+	TEST_ASSERT_VAL("wrong nr", map->nr == 1);
+	TEST_ASSERT_VAL("wrong pid",
+			perf_thread_map__pid(map, 0) == getpid());
+	TEST_ASSERT_VAL("wrong comm",
+			perf_thread_map__comm(map, 0) &&
+			!strcmp(perf_thread_map__comm(map, 0), NAME));
+	TEST_ASSERT_VAL("wrong refcnt",
+			refcount_read(&map->refcnt) == 1);
+	perf_thread_map__put(map);
+
+	/* test dummy pid */
+	map = perf_thread_map__new_dummy();
+	TEST_ASSERT_VAL("failed to alloc map", map);
+
+	thread_map__read_comms(map);
+
+	TEST_ASSERT_VAL("wrong nr", map->nr == 1);
+	TEST_ASSERT_VAL("wrong pid", perf_thread_map__pid(map, 0) == -1);
+	TEST_ASSERT_VAL("wrong comm",
+			perf_thread_map__comm(map, 0) &&
+			!strcmp(perf_thread_map__comm(map, 0), "dummy"));
+	TEST_ASSERT_VAL("wrong refcnt",
+			refcount_read(&map->refcnt) == 1);
+	perf_thread_map__put(map);
+	return 0;
+}
+
+static int process_event(const struct perf_tool *tool __maybe_unused,
+			 union perf_event *event,
+			 struct perf_sample *sample __maybe_unused,
+			 struct machine *machine __maybe_unused)
+{
+	struct perf_record_thread_map *map = &event->thread_map;
+	struct perf_thread_map *threads;
+
+	TEST_ASSERT_VAL("wrong nr",   map->nr == 1);
+	TEST_ASSERT_VAL("wrong pid",  map->entries[0].pid == (u64) getpid());
+	TEST_ASSERT_VAL("wrong comm", !strcmp(map->entries[0].comm, NAME));
+
+	threads = thread_map__new_event(&event->thread_map);
+	TEST_ASSERT_VAL("failed to alloc map", threads);
+
+	TEST_ASSERT_VAL("wrong nr", threads->nr == 1);
+	TEST_ASSERT_VAL("wrong pid",
+			perf_thread_map__pid(threads, 0) == getpid());
+	TEST_ASSERT_VAL("wrong comm",
+			perf_thread_map__comm(threads, 0) &&
+			!strcmp(perf_thread_map__comm(threads, 0), NAME));
+	TEST_ASSERT_VAL("wrong refcnt",
+			refcount_read(&threads->refcnt) == 1);
+	perf_thread_map__put(threads);
+	return 0;
+}
+
+static int test__thread_map_synthesize(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
+{
+	struct perf_thread_map *threads;
+
+	TEST_ASSERT_VAL("failed to set process name",
+			!prctl(PR_SET_NAME, NAMEUL, 0, 0, 0));
+
+	/* test map on current pid */
+	threads = thread_map__new_by_pid(getpid());
+	TEST_ASSERT_VAL("failed to alloc map", threads);
+
+	thread_map__read_comms(threads);
+
+	TEST_ASSERT_VAL("failed to synthesize map",
+		!perf_event__synthesize_thread_map2(NULL, threads, process_event, NULL));
+
+	perf_thread_map__put(threads);
+	return 0;
+}
+
+static int test__thread_map_remove(struct test_suite *test __maybe_unused, int subtest __maybe_unused)
+{
+	struct perf_thread_map *threads;
+	char *str;
+
+	TEST_ASSERT_VAL("failed to allocate map string",
+			asprintf(&str, "%d,%d", getpid(), getppid()) >= 0);
+
+	threads = thread_map__new_str(str, /*tid=*/NULL, /*all_threads=*/false);
+	free(str);
+
+	TEST_ASSERT_VAL("failed to allocate thread_map",
+			threads);
+
+	if (verbose > 0)
+		thread_map__fprintf(threads, stderr);
+
+	TEST_ASSERT_VAL("failed to remove thread",
+			!thread_map__remove(threads, 0));
+
+	TEST_ASSERT_VAL("thread_map count != 1", threads->nr == 1);
+
+	if (verbose > 0)
+		thread_map__fprintf(threads, stderr);
+
+	TEST_ASSERT_VAL("failed to remove thread",
+			!thread_map__remove(threads, 0));
+
+	TEST_ASSERT_VAL("thread_map count != 0", threads->nr == 0);
+
+	if (verbose > 0)
+		thread_map__fprintf(threads, stderr);
+
+	TEST_ASSERT_VAL("failed to not remove thread",
+			thread_map__remove(threads, 0));
+
+	perf_thread_map__put(threads);
+	return 0;
+}
+
+DEFINE_SUITE("Thread map", thread_map);
+DEFINE_SUITE("Synthesize thread map", thread_map_synthesize);
+DEFINE_SUITE("Remove thread map", thread_map_remove);