[PATCH 0/2] cgroup, sched: add bpf for cgroup cpu controller

[Ziyang Men <ziyang.meme@gmail.com>]

As the title suggests, this series adds bpf kfuncs for the cgroup CPU
controller.

Motivation:
Collecting cgroup statistics is expensive: the existing
method is to open and parse a cgroup file. memcg already has an
efficient alternative through BPF; this series extends that idea to cpu.

Design:
- Leave reading the CFS bandwidth counters to the BPF program. They are
  plain fields of tg->cfs_bandwidth, so they need no kernel code.
- Add one kfunc to compute the throttled time. This is necessary because
  it is a sum over every possible cpu, which a user cannot do itself.

The only part it touches the scheduler part is to discard the static for
throttled_time_self() in order to use externally.

Patch 1 adds the kfuncs (kernel/cgroup/bpf_cpu.c); patch 2 adds a
test_progs selftest.

Tested on VM with v7.2-rc5. 

Ziyang Men (2):
  cgroup, sched: add BPF kfuncs to read a cpu cgroup's stats
  selftests/bpf: add cgroup_iter_cpu test for cpu cgroup kfuncs

 include/linux/cgroup.h                        |  15 +
 kernel/cgroup/Makefile                        |   2 +
 kernel/cgroup/bpf_cpu.c                       |  80 ++++++
 kernel/cgroup/cgroup-internal.h               |   3 +
 kernel/cgroup/rstat.c                         |  42 +++
 kernel/sched/core.c                           |   2 +-
 tools/testing/selftests/bpf/cgroup_iter_cpu.h |  22 ++
 tools/testing/selftests/bpf/config            |   3 +
 .../bpf/prog_tests/cgroup_iter_cpu.c          | 259 ++++++++++++++++++
 .../selftests/bpf/progs/cgroup_iter_cpu.c     |  53 ++++
 10 files changed, 480 insertions(+), 1 deletion(-)
 create mode 100644 kernel/cgroup/bpf_cpu.c
 create mode 100644 tools/testing/selftests/bpf/cgroup_iter_cpu.h
 create mode 100644 tools/testing/selftests/bpf/prog_tests/cgroup_iter_cpu.c
 create mode 100644 tools/testing/selftests/bpf/progs/cgroup_iter_cpu.c

-- 
2.53.0-Meta