Received-SPF: softfail (google.com: domain of transitioning linux-kernel-owner@vger.kernel.org does not designate 23.128.96.19 as permitted sender) client-ip=23.128.96.19;
MIME-Version: 1.0
References: <20220312171354.661448-1-ytcoode@gmail.com>
In-Reply-To: <20220312171354.661448-1-ytcoode@gmail.com>
From:   Andrii Nakryiko <andrii.nakryiko@gmail.com>
Date:   Tue, 15 Mar 2022 11:36:56 -0700
Message-ID: <CAEf4BzbzqvQM63-mO96tbNaPXsKSbff4h-mX6UBfbU9zZG67OQ@mail.gmail.com>
Subject: Re: [PATCH bpf-next] libbpf: Remove redundant check in btf_ext__new()
To:     Yuntao Wang <ytcoode@gmail.com>
Cc:     Alexei Starovoitov <ast@kernel.org>,
        Daniel Borkmann <daniel@iogearbox.net>,
        Andrii Nakryiko <andrii@kernel.org>,
        Martin KaFai Lau <kafai@fb.com>,
        Song Liu <songliubraving@fb.com>, Yonghong Song <yhs@fb.com>,
        John Fastabend <john.fastabend@gmail.com>,
        KP Singh <kpsingh@kernel.org>,
        Networking <netdev@vger.kernel.org>, bpf <bpf@vger.kernel.org>,
        open list <linux-kernel@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Sat, Mar 12, 2022 at 9:14 AM Yuntao Wang <ytcoode@gmail.com> wrote:
>
> Since 'core_relo_len' is the last field of 'struct btf_ext_header', if
> 'xxx->hdr_len' is not less than 'offsetofend(xxx, core_relo_len)', then
> 'xxx->hdr_len' must also be not less than 'offsetofend(xxx, line_info_len)'.
>
> We can check 'xxx->hdr_len < offsetofend(xxx, core_relo_len)' first, if it
> passes, the 'xxx->hdr_len < offsetofend(xxx, line_info_len)' check will be
> redundant, it can be removed.
>
> Signed-off-by: Yuntao Wang <ytcoode@gmail.com>
> ---
>  tools/lib/bpf/btf.c | 7 +------
>  1 file changed, 1 insertion(+), 6 deletions(-)
>
> diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c
> index 1383e26c5d1f..d55b44124c3e 100644
> --- a/tools/lib/bpf/btf.c
> +++ b/tools/lib/bpf/btf.c
> @@ -2813,7 +2813,7 @@ struct btf_ext *btf_ext__new(const __u8 *data, __u32 size)
>         if (err)
>                 goto done;
>
> -       if (btf_ext->hdr->hdr_len < offsetofend(struct btf_ext_header, line_info_len)) {
> +       if (btf_ext->hdr->hdr_len < offsetofend(struct btf_ext_header, core_relo_len)) {
>                 err = -EINVAL;
>                 goto done;
>         }
> @@ -2826,11 +2826,6 @@ struct btf_ext *btf_ext__new(const __u8 *data, __u32 size)
>         if (err)
>                 goto done;
>
> -       if (btf_ext->hdr->hdr_len < offsetofend(struct btf_ext_header, core_relo_len)) {
> -               err = -EINVAL;
> -               goto done;
> -       }

it seems like it's actually a bug. If header is smaller then core
relos parsing should be skipped, I think. Maybe let's fix that
instead?

basically the logic should be:

1. if size of header is exactly == offsetof(core_relo_off) then skip core relos
2. otherwise check that it has enough size to cover core_relo_off and
core_relo_len, and error out if not
3. otherwise proceed to parsing core relos

> -
>         err = btf_ext_setup_core_relos(btf_ext);
>         if (err)
>                 goto done;
> --
> 2.35.1
>