memcg: reading memcg stats more efficiently

[kernel-patches-daemon-bpf-rc]

Pull request for series with
subject: memcg: reading memcg stats more efficiently
version: 2
url: https://patchwork.kernel.org/project/netdevbpf/list/?series=1012038

[kernel-patches-daemon-bpf-rc]

Upstream branch: 48a97ff
series: https://patchwork.kernel.org/project/netdevbpf/list/?series=1012038
version: 2

[kernel-patches-review-bot]

commit 141a6230950e8b781e9438b1d3031107f7ab2155
Author: JP Kobryn <[email protected]>

memcg: introduce kfuncs for fetching memcg stats

This patch introduces four BPF kfuncs for fetching memory cgroup statistics
directly from BPF programs, avoiding the overhead of reading and parsing the
memory.stat file. The kfuncs provide access to memcg stats, node stats, and
vm events.

> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> index 4deda3362..6547c27d4 100644
> --- a/mm/memcontrol.c
> +++ b/mm/memcontrol.c
> @@ -871,6 +871,73 @@ unsigned long memcg_events_local(struct mem_cgroup *memcg, int event)
>  }
>  #endif
>
> +static inline struct mem_cgroup *memcg_from_cgroup(struct cgroup *cgrp)
> +{
> +	return cgrp ? mem_cgroup_from_css(cgrp->subsys[memory_cgrp_id]) : NULL;
                                              ^^^^^^
The subsys array is marked __rcu in struct cgroup (see
include/linux/cgroup-defs.h:544). Other code accessing cgrp->subsys[]
uses rcu_dereference_check() or rcu_dereference_raw() (see
kernel/cgroup/cgroup.c:531 and kernel/cgroup/cgroup.c:703).

Should this access use an RCU dereference primitive? If BPF programs
execute within an RCU read-side critical section, it would be good to
make that explicit here for consistency with the rest of the codebase.

> +}
> +
> +__bpf_kfunc static void memcg_flush_stats(struct cgroup *cgrp)
> +{
> +	struct mem_cgroup *memcg = memcg_from_cgroup(cgrp);
> +
> +	if (!memcg)
> +		return;
> +
> +	mem_cgroup_flush_stats(memcg);
> +}
> +
> +__bpf_kfunc static unsigned long memcg_stat_fetch(struct cgroup *cgrp,
> +		enum memcg_stat_item item)
> +{
> +	struct mem_cgroup *memcg = memcg_from_cgroup(cgrp);
> +
> +	if (!memcg)
> +		return 0;
> +
> +	return memcg_page_state_output(memcg, item);
> +}
> +
> +__bpf_kfunc static unsigned long memcg_node_stat_fetch(struct cgroup *cgrp,
> +		enum node_stat_item item)
> +{
> +	struct mem_cgroup *memcg = memcg_from_cgroup(cgrp);
> +
> +	if (!memcg)
> +		return 0;
> +
> +	return memcg_page_state_output(memcg, item);
> +}
> +
> +__bpf_kfunc static unsigned long memcg_vm_event_fetch(struct cgroup *cgrp,
> +		enum vm_event_item item)
> +{
> +	struct mem_cgroup *memcg = memcg_from_cgroup(cgrp);
> +
> +	if (!memcg)
> +		return 0;
> +
> +	return memcg_events(memcg, item);
> +}
> +
> +BTF_KFUNCS_START(bpf_memcontrol_kfunc_ids)
> +BTF_ID_FLAGS(func, memcg_flush_stats, KF_TRUSTED_ARGS | KF_SLEEPABLE)
> +BTF_ID_FLAGS(func, memcg_stat_fetch, KF_TRUSTED_ARGS)
> +BTF_ID_FLAGS(func, memcg_node_stat_fetch, KF_TRUSTED_ARGS)
> +BTF_ID_FLAGS(func, memcg_vm_event_fetch, KF_TRUSTED_ARGS)
> +BTF_KFUNCS_END(bpf_memcontrol_kfunc_ids)
> +
> +static const struct btf_kfunc_id_set bpf_memcontrol_kfunc_set = {
> +	.owner          = THIS_MODULE,
> +	.set            = &bpf_memcontrol_kfunc_ids,
> +};
> +
> +static int __init bpf_memcontrol_kfunc_init(void)
> +{
> +	return register_btf_kfunc_id_set(BPF_PROG_TYPE_UNSPEC,
> +					 &bpf_memcontrol_kfunc_set);
> +}
> +late_initcall(bpf_memcontrol_kfunc_init);
> +
>  struct mem_cgroup *mem_cgroup_from_task(struct task_struct *p)
>  {
>  	/*

---

AI reviewed your patch. Please fix the bug or email reply why it's not a bug.
See: https://github.com/kernel-patches/vmtest/blob/master/ci/claude/README.md

In-Reply-To-Subject: memcg: introduce kfuncs for fetching memcg stats
CI run summary: https://github.com/kernel-patches/bpf-rc/actions/runs/18539971331

[kernel-patches-daemon-bpf-rc]

Forwarding comment 3407981460 via email
In-Reply-To: [email protected]
Patch: https://patchwork.kernel.org/project/netdevbpf/patch/[email protected]/

[kernel-patches-daemon-bpf-rc]

Upstream branch: 48a97ff
series: https://patchwork.kernel.org/project/netdevbpf/list/?series=1012038
version: 2

[kernel-patches-daemon-bpf-rc]

Upstream branch: 48a97ff
series: https://patchwork.kernel.org/project/netdevbpf/list/?series=1012038
version: 2

[kernel-patches-daemon-bpf-rc]

Upstream branch: 48a97ff
series: https://patchwork.kernel.org/project/netdevbpf/list/?series=1012038
version: 2

[kernel-patches-daemon-bpf-rc]

At least one diff in series https://patchwork.kernel.org/project/netdevbpf/list/?series=1012038 expired. Closing PR.