Subject: Re: [PATCH net-next 4/5] bpftool: implement cgdetach command
To: Roman Gushchin <guro@fb.com>, netdev@vger.kernel.org
Cc: linux-kernel@vger.kernel.org, kernel-team@fb.com, ast@kernel.org,
        daniel@iogearbox.net, jakub.kicinski@netronome.com, kafai@fb.com
References: <20171130134302.2840-1-guro@fb.com>
 <20171130134302.2840-5-guro@fb.com>
From: Quentin Monnet <quentin.monnet@netronome.com>
Message-ID: <150c6e31-2fdb-3448-d434-773ebd24e351@netronome.com>
Date: Fri, 1 Dec 2017 10:26:27 +0000
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101
 Thunderbird/52.5.0
MIME-Version: 1.0
In-Reply-To: <20171130134302.2840-5-guro@fb.com>
Content-Type: text/plain; charset=utf-8
Content-Language: en-US
Content-Transfer-Encoding: 8bit
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 2240
Lines: 86

2017-11-30 13:43 UTC+0000 ~ Roman Gushchin <guro@fb.com>
> Implement cgdetach command, which allows to detach the bpf
> program from a cgroup. It takes program id and attach type
> as arguments.
> 
> Example:
> $ ./bpftool cgdetach /sys/fs/cgroup/user.slice/ device 1
> 
> Signed-off-by: Roman Gushchin <guro@fb.com>
> Cc: Alexei Starovoitov <ast@kernel.org>
> Cc: Daniel Borkmann <daniel@iogearbox.net>
> Cc: Jakub Kicinski <jakub.kicinski@netronome.com>
> Cc: Martin KaFai Lau <kafai@fb.com>
> ---
>  tools/bpf/bpftool/main.c | 50 +++++++++++++++++++++++++++++++++++++++++++++++-
>  1 file changed, 49 insertions(+), 1 deletion(-)
> 
> diff --git a/tools/bpf/bpftool/main.c b/tools/bpf/bpftool/main.c
> index 8eb3b9bf5bb2..77fcc1a0bd5d 100644
> --- a/tools/bpf/bpftool/main.c
> +++ b/tools/bpf/bpftool/main.c

[…]

> @@ -338,6 +341,51 @@ static int do_cgattach(int argc, char **argv)
>  	return 0;
>  }
>  
> +static int do_cgdetach(int argc, char **argv)
> +{
> +	int prog_fd, cgroup_fd;
> +	enum bpf_attach_type attach_type;
> +
> +	if (argc < 3) {
> +		p_err("too few parameters for cgdetach\n");
> +		return -1;
> +	} else if (argc > 3) {
> +		p_err("too many parameters for cgdetach\n");
> +		return -1;
> +	}
> +
> +	cgroup_fd = open(argv[0], O_RDONLY);
> +	if (cgroup_fd < 0) {
> +		p_err("can't open cgroup %s\n", argv[1]);
> +		return -1;
> +	}
> +
> +	attach_type = parse_attach_type(argv[1]);
> +	if (attach_type == __MAX_BPF_ATTACH_TYPE) {
> +		close(cgroup_fd);
> +		p_err("Invalid attach type");
> +		return -1;
> +	}
> +
> +	prog_fd = bpf_prog_get_fd_by_id(atoi(argv[2]));
> +	if (prog_fd < 0) {
> +		p_err("invalid program id\n");
> +		return -1;
> +	}
> +
> +	if (bpf_prog_detach2(prog_fd, cgroup_fd, attach_type)) {
> +		close(prog_fd);
> +		close(cgroup_fd);
> +		p_err("Failed to attach program");
> +		return -1;
> +	}
> +
> +	close(prog_fd);
> +	close(cgroup_fd);

Could you please make the function print a "null" string for JSON here?
So that it does not break JSON for batched commands. Should be as simple as:

	jsonw_null(json_wtr);

This is also valid for `do_cgattach()` in patch 3.
Thanks!

> +
> +	return 0;
> +}
> +
>  int main(int argc, char **argv)
>  {
>  	static const struct option options[] = {
>