]> git.ipfire.org Git - thirdparty/kernel/linux.git/commit
mm: introduce bpf_get_root_mem_cgroup() BPF kfunc
authorRoman Gushchin <roman.gushchin@linux.dev>
Tue, 23 Dec 2025 04:41:53 +0000 (20:41 -0800)
committerAlexei Starovoitov <ast@kernel.org>
Tue, 23 Dec 2025 06:20:21 +0000 (22:20 -0800)
commit5c7db3239c9fbe3c62cb0d89b64959ea23af2de9
treea2bff5530dfeef1d9795ce009f42c84d4ab42172
parent5904db9891f80f84283648121e2d8c8a506296a8
mm: introduce bpf_get_root_mem_cgroup() BPF kfunc

Introduce a BPF kfunc to get a trusted pointer to the root memory
cgroup. It's very handy to traverse the full memcg tree, e.g.
for handling a system-wide OOM.

It's possible to obtain this pointer by traversing the memcg tree
up from any known memcg, but it's sub-optimal and makes BPF programs
more complex and less efficient.

bpf_get_root_mem_cgroup() has a KF_ACQUIRE | KF_RET_NULL semantics,
however in reality it's not necessary to bump the corresponding
reference counter - root memory cgroup is immortal, reference counting
is skipped, see css_get(). Once set, root_mem_cgroup is always a valid
memcg pointer. It's safe to call bpf_put_mem_cgroup() for the pointer
obtained with bpf_get_root_mem_cgroup(), it's effectively a no-op.

Signed-off-by: Roman Gushchin <roman.gushchin@linux.dev>
Link: https://lore.kernel.org/r/20251223044156.208250-4-roman.gushchin@linux.dev
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
mm/bpf_memcontrol.c