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, 92 insertions, 0 deletions

diff --git a/tools/perf/util/cloexec.c b/tools/perf/util/cloexec.c
new file mode 100644
index 000000000..8830604c3
--- /dev/null
+++ b/tools/perf/util/cloexec.c
@@ -0,0 +1,92 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <errno.h>
+#include <sched.h>
+#include "util.h" // for sched_getcpu()
+#include "../perf-sys.h"
+#include "cloexec.h"
+#include "event.h"
+#include "asm/bug.h"
+#include "debug.h"
+#include <unistd.h>
+#include <sys/syscall.h>
+#include <linux/string.h>
+
+static unsigned long flag = PERF_FLAG_FD_CLOEXEC;
+
+static int perf_flag_probe(void)
+{
+	/* use 'safest' configuration as used in evsel__fallback() */
+	struct perf_event_attr attr = {
+		.type = PERF_TYPE_SOFTWARE,
+		.config = PERF_COUNT_SW_CPU_CLOCK,
+		.exclude_kernel = 1,
+	};
+	int fd;
+	int err;
+	int cpu;
+	pid_t pid = -1;
+	char sbuf[STRERR_BUFSIZE];
+
+	cpu = sched_getcpu();
+	if (cpu < 0)
+		cpu = 0;
+
+	/*
+	 * Using -1 for the pid is a workaround to avoid gratuitous jump label
+	 * changes.
+	 */
+	while (1) {
+		/* check cloexec flag */
+		fd = sys_perf_event_open(&attr, pid, cpu, -1,
+					 PERF_FLAG_FD_CLOEXEC);
+		if (fd < 0 && pid == -1 && errno == EACCES) {
+			pid = 0;
+			continue;
+		}
+		break;
+	}
+	err = errno;
+
+	if (fd >= 0) {
+		close(fd);
+		return 1;
+	}
+
+	WARN_ONCE(err != EINVAL && err != EBUSY && err != EACCES,
+		  "perf_event_open(..., PERF_FLAG_FD_CLOEXEC) failed with unexpected error %d (%s)\n",
+		  err, str_error_r(err, sbuf, sizeof(sbuf)));
+
+	/* not supported, confirm error related to PERF_FLAG_FD_CLOEXEC */
+	while (1) {
+		fd = sys_perf_event_open(&attr, pid, cpu, -1, 0);
+		if (fd < 0 && pid == -1 && errno == EACCES) {
+			pid = 0;
+			continue;
+		}
+		break;
+	}
+	err = errno;
+
+	if (fd >= 0)
+		close(fd);
+
+	if (WARN_ONCE(fd < 0 && err != EBUSY && err != EACCES,
+		      "perf_event_open(..., 0) failed unexpectedly with error %d (%s)\n",
+		      err, str_error_r(err, sbuf, sizeof(sbuf))))
+		return -1;
+
+	return 0;
+}
+
+unsigned long perf_event_open_cloexec_flag(void)
+{
+	static bool probed;
+
+	if (!probed) {
+		if (perf_flag_probe() <= 0)
+			flag = 0;
+		probed = true;
+	}
+
+	return flag;
+}