On Mon, Aug 1, 2022 at 3:00 PM Andrii Nakryiko
[off-list ref] wrote:

On Mon, Aug 1, 2022 at 10:54 AM Hao Luo [off-list ref] wrote:

quoted

[...]

quoted

---
 .../prog_tests/cgroup_hierarchical_stats.c    | 358 ++++++++++++++++++
 .../bpf/progs/cgroup_hierarchical_stats.c     | 218 +++++++++++
 2 files changed, 576 insertions(+)
 create mode 100644 tools/testing/selftests/bpf/prog_tests/cgroup_hierarchical_stats.c
 create mode 100644 tools/testing/selftests/bpf/progs/cgroup_hierarchical_stats.c

[...]

quoted

+extern void cgroup_rstat_updated(struct cgroup *cgrp, int cpu) __ksym;
+extern void cgroup_rstat_flush(struct cgroup *cgrp) __ksym;
+
+static struct cgroup *task_memcg(struct task_struct *task)
+{
+       return task->cgroups->subsys[memory_cgrp_id]->cgroup;

memory_cgrp_id is kernel-defined internal enum which actually can
change based on kernel configuration (i.e., which cgroup subsystems
are enabled or not), is that right?

In practice you wouldn't hard-code it, it's better to use
bpf_core_enum_value() to capture enum's value in CO-RE-relocatable
way.

So it might be a good idea to demonstrate that here.

Makes sense. Will use bpf_core_enum_value. Thanks!

quoted

+}
+
+static uint64_t cgroup_id(struct cgroup *cgrp)
+{
+       return cgrp->kn->id;
+}
+

[...]