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, 209 insertions, 0 deletions

diff --git a/tools/perf/util/stat.h b/tools/perf/util/stat.h
new file mode 100644
index 000000000..4bced233d
--- /dev/null
+++ b/tools/perf/util/stat.h
@@ -0,0 +1,209 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef __PERF_STATS_H
+#define __PERF_STATS_H
+
+#include <linux/types.h>
+#include <stdio.h>
+#include <sys/types.h>
+#include <sys/resource.h>
+#include "cpumap.h"
+#include "counts.h"
+
+struct perf_cpu_map;
+struct perf_stat_config;
+struct timespec;
+
+struct stats {
+	double n, mean, M2;
+	u64 max, min;
+};
+
+/* hold aggregated event info */
+struct perf_stat_aggr {
+	/* aggregated values */
+	struct perf_counts_values	counts;
+	/* number of entries (CPUs) aggregated */
+	int				nr;
+	/* whether any entry has failed to read/process event */
+	bool				failed;
+	/* to mark this data is processed already */
+	bool				used;
+};
+
+/* per-evsel event stats */
+struct perf_stat_evsel {
+	/* used for repeated runs */
+	struct stats		 res_stats;
+	/* number of allocated 'aggr' */
+	int			 nr_aggr;
+	/* aggregated event values */
+	struct perf_stat_aggr	*aggr;
+	/* used for group read */
+	u64			*group_data;
+};
+
+enum aggr_mode {
+	AGGR_NONE,
+	AGGR_GLOBAL,
+	AGGR_SOCKET,
+	AGGR_DIE,
+	AGGR_CLUSTER,
+	AGGR_CACHE,
+	AGGR_CORE,
+	AGGR_THREAD,
+	AGGR_UNSET,
+	AGGR_NODE,
+	AGGR_MAX
+};
+
+typedef struct aggr_cpu_id (*aggr_get_id_t)(struct perf_stat_config *config, struct perf_cpu cpu);
+
+struct perf_stat_config {
+	enum aggr_mode		 aggr_mode;
+	u32			 aggr_level;
+	bool			 scale;
+	bool			 no_inherit;
+	bool			 identifier;
+	bool			 csv_output;
+	bool			 json_output;
+	bool			 interval_clear;
+	bool			 metric_only;
+	bool			 null_run;
+	bool			 ru_display;
+	bool			 big_num;
+	bool			 hybrid_merge;
+	bool			 walltime_run_table;
+	bool			 all_kernel;
+	bool			 all_user;
+	bool			 percore_show_thread;
+	bool			 summary;
+	bool			 no_csv_summary;
+	bool			 metric_no_group;
+	bool			 metric_no_merge;
+	bool			 metric_no_threshold;
+	bool			 hardware_aware_grouping;
+	bool			 stop_read_counter;
+	bool			 iostat_run;
+	char			 *user_requested_cpu_list;
+	bool			 system_wide;
+	FILE			*output;
+	unsigned int		 interval;
+	unsigned int		 timeout;
+	unsigned int		 unit_width;
+	unsigned int		 metric_only_len;
+	int			 times;
+	int			 run_count;
+	int			 print_free_counters_hint;
+	const char		*csv_sep;
+	struct stats		*walltime_nsecs_stats;
+	struct rusage		 ru_data;
+	struct cpu_aggr_map	*aggr_map;
+	aggr_get_id_t		 aggr_get_id;
+	struct cpu_aggr_map	*cpus_aggr_map;
+	u64			*walltime_run;
+	int			 ctl_fd;
+	int			 ctl_fd_ack;
+	bool			 ctl_fd_close;
+	const char		*cgroup_list;
+	unsigned int		topdown_level;
+};
+
+extern struct perf_stat_config stat_config;
+
+void perf_stat__set_big_num(int set);
+
+void update_stats(struct stats *stats, u64 val);
+double avg_stats(struct stats *stats);
+double stddev_stats(struct stats *stats);
+double rel_stddev_stats(double stddev, double avg);
+
+static inline void init_stats(struct stats *stats)
+{
+	stats->n    = 0.0;
+	stats->mean = 0.0;
+	stats->M2   = 0.0;
+	stats->min  = (u64) -1;
+	stats->max  = 0;
+}
+
+struct evsel;
+struct evlist;
+
+enum metric_threshold_classify {
+	METRIC_THRESHOLD_UNKNOWN,
+	METRIC_THRESHOLD_BAD,
+	METRIC_THRESHOLD_NEARLY_BAD,
+	METRIC_THRESHOLD_LESS_GOOD,
+	METRIC_THRESHOLD_GOOD,
+};
+const char *metric_threshold_classify__color(enum metric_threshold_classify thresh);
+
+typedef void (*print_metric_t)(struct perf_stat_config *config,
+			       void *ctx,
+			       enum metric_threshold_classify thresh,
+			       const char *fmt,
+			       const char *unit,
+			       double val);
+typedef void (*new_line_t)(struct perf_stat_config *config, void *ctx);
+
+/* Used to print the display name of the Default metricgroup for now. */
+typedef void (*print_metricgroup_header_t)(struct perf_stat_config *config,
+					   void *ctx, const char *metricgroup_name);
+
+void perf_stat__reset_shadow_stats(void);
+struct perf_stat_output_ctx {
+	void *ctx;
+	print_metric_t print_metric;
+	new_line_t new_line;
+	print_metricgroup_header_t print_metricgroup_header;
+	bool force_header;
+};
+
+void perf_stat__print_shadow_stats(struct perf_stat_config *config,
+				   struct evsel *evsel,
+				   int aggr_idx,
+				   struct perf_stat_output_ctx *out);
+bool perf_stat__skip_metric_event(struct evsel *evsel);
+void *perf_stat__print_shadow_stats_metricgroup(struct perf_stat_config *config,
+						struct evsel *evsel,
+						int aggr_idx,
+						int *num,
+						void *from,
+						struct perf_stat_output_ctx *out);
+
+int evlist__alloc_stats(struct perf_stat_config *config,
+			struct evlist *evlist, bool alloc_raw);
+void evlist__free_stats(struct evlist *evlist);
+void evlist__reset_stats(struct evlist *evlist);
+void evlist__reset_prev_raw_counts(struct evlist *evlist);
+void evlist__copy_prev_raw_counts(struct evlist *evlist);
+void evlist__save_aggr_prev_raw_counts(struct evlist *evlist);
+
+int evlist__alloc_aggr_stats(struct evlist *evlist, int nr_aggr);
+void evlist__reset_aggr_stats(struct evlist *evlist);
+void evlist__copy_res_stats(struct perf_stat_config *config, struct evlist *evlist);
+
+int perf_stat_process_counter(struct perf_stat_config *config,
+			      struct evsel *counter);
+void perf_stat_merge_counters(struct perf_stat_config *config, struct evlist *evlist);
+void perf_stat_process_percore(struct perf_stat_config *config, struct evlist *evlist);
+
+struct perf_tool;
+union perf_event;
+struct perf_session;
+struct target;
+
+int perf_event__process_stat_event(const struct perf_tool *tool,
+				   struct perf_session *session,
+				   union perf_event *event);
+
+size_t perf_event__fprintf_stat(union perf_event *event, FILE *fp);
+size_t perf_event__fprintf_stat_round(union perf_event *event, FILE *fp);
+size_t perf_event__fprintf_stat_config(union perf_event *event, FILE *fp);
+
+void evlist__print_counters(struct evlist *evlist, struct perf_stat_config *config,
+			    struct target *_target, struct timespec *ts, int argc, const char **argv);
+
+struct metric_expr;
+double test_generic_metric(struct metric_expr *mexp, int aggr_idx);
+#endif