Re: [PATCH bpf-next v5 4/8] bpf: Introduce cgroup iter

[Tejun Heo <tj@kernel.org>]

Hello,

On Fri, Jul 22, 2022 at 05:48:25PM +0000, Yosry Ahmed wrote:
> +
> +	/* cgroup_iter walks either the live descendants of a cgroup subtree, or the
> +	 * ancestors of a given cgroup.
> +	 */
> +	struct {
> +		/* Cgroup file descriptor. This is root of the subtree if walking
> +		 * descendants; it's the starting cgroup if walking the ancestors.
> +		 * If it is left 0, the traversal starts from the default cgroup v2
> +		 * root. For walking v1 hierarchy, one should always explicitly
> +		 * specify the cgroup_fd.
> +		 */
> +		__u32	cgroup_fd;

So, we're identifying the starting point with an fd.

> +		__u32	traversal_order;
> +	} cgroup;
>  };
>  
>  /* BPF syscall commands, see bpf(2) man-page for more details. */
> @@ -6136,6 +6156,16 @@ struct bpf_link_info {
>  					__u32 map_id;
>  				} map;
>  			};
> +			union {
> +				struct {
> +					__u64 cgroup_id;
> +					__u32 traversal_order;
> +				} cgroup;

but iterating the IDs. IDs are the better choice for cgroup2 as there's
nothing specific to the calling program or the fds it has, but I guess this
is because you want to use it for cgroup1, right? Oh well, that's okay I
guess.

Acked-by: Tejun Heo <tj@kernel.org>

Thanks.

-- 
tejun