Lines Matching full:leader
367 if (evsel->core.leader->nr_members > 1) { in bperf_check_target()
403 pr_err("Failed to open leader skeleton\n"); in bperf_reload_leader_program()
410 pr_err("Failed to load leader skeleton\n"); in bperf_reload_leader_program()
416 pr_err("Failed to attach leader program\n"); in bperf_reload_leader_program()
464 * Step 1: hold a fd on the leader program and the bpf_link, if in bperf__load()
489 * The bpf_link holds reference to the leader program, and the in bperf__load()
490 * leader program holds reference to the maps. Therefore, if in bperf__load()
519 /* attach fexit program to the leader program */ in bperf__load()
523 /* connect to leader diff_reading map */ in bperf__load()
583 * trigger the leader prog on each cpu, so the accum_reading map could get
710 * cs -> | leader prog |/ ||/ | |
730 * The leader prog is triggered on each context switch (cs). The leader
733 * multiple perf-stat sessions share the same leader prog.
736 * leader prog. It is possible to attach up to BPF_MAX_TRAMP_PROGS (38)
737 * follower progs to the same leader prog. The follower prog checks current
743 * Besides context switch, it is also necessary to trigger the leader prog
750 * the leader prog.