netdev - Re: [PATCH net v4] bpf, net: Check cgroup_bpf_enabled() only once in do_sock

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAADnVQLLN9hbQ8FQnX_uWFAVBd7L9HhsQpQymLOmB-dHFR4VRw@mail.gmail.com>
Date: Wed, 21 Aug 2024 14:01:08 -0700
From: Alexei Starovoitov <alexei.starovoitov@...il.com>
To: Tze-nan Wu <Tze-nan.Wu@...iatek.com>
Cc: Network Development <netdev@...r.kernel.org>, bpf <bpf@...r.kernel.org>, 
	"David S. Miller" <davem@...emloft.net>, Eric Dumazet <edumazet@...gle.com>, 
	Jakub Kicinski <kuba@...nel.org>, Paolo Abeni <pabeni@...hat.com>, 
	Matthias Brugger <matthias.bgg@...il.com>, 
	AngeloGioacchino Del Regno <angelogioacchino.delregno@...labora.com>, 
	Stanislav Fomichev <sdf@...ichev.me>, bobule.chang@...iatek.com, wsd_upstream@...iatek.com, 
	LKML <linux-kernel@...r.kernel.org>, linux-mediatek@...ts.infradead.org, 
	Kuniyuki Iwashima <kuniyu@...zon.com>, Yanghui Li <yanghui.li@...iatek.com>, 
	Cheng-Jui Wang <cheng-jui.wang@...iatek.com>, Alexei Starovoitov <ast@...nel.org>, 
	Daniel Borkmann <daniel@...earbox.net>, John Fastabend <john.fastabend@...il.com>, 
	Andrii Nakryiko <andrii@...nel.org>, Martin KaFai Lau <martin.lau@...ux.dev>, 
	Eduard Zingerman <eddyz87@...il.com>, Song Liu <song@...nel.org>, 
	Yonghong Song <yonghong.song@...ux.dev>, KP Singh <kpsingh@...nel.org>, 
	Hao Luo <haoluo@...gle.com>, Jiri Olsa <jolsa@...nel.org>, 
	linux-arm-kernel <linux-arm-kernel@...ts.infradead.org>
Subject: Re: [PATCH net v4] bpf, net: Check cgroup_bpf_enabled() only once in do_sock_getsockopt()

On Wed, Aug 21, 2024 at 2:30 AM Tze-nan Wu <Tze-nan.Wu@...iatek.com> wrote:
>
> The return value from `cgroup_bpf_enabled(CGROUP_GETSOCKOPT)` can change
> between the invocations of `BPF_CGROUP_GETSOCKOPT_MAX_OPTLEN` and
> `BPF_CGROUP_RUN_PROG_GETSOCKOPT`.
>
> If `cgroup_bpf_enabled(CGROUP_GETSOCKOPT)` changes from "false" to
> "true" between the invocations of `BPF_CGROUP_GETSOCKOPT_MAX_OPTLEN` and
> `BPF_CGROUP_RUN_PROG_GETSOCKOPT`, `BPF_CGROUP_RUN_PROG_GETSOCKOPT` will
> receive an -EFAULT from `__cgroup_bpf_run_filter_getsockopt(max_optlen=0)`
> due to `get_user()` was not reached in `BPF_CGROUP_GETSOCKOPT_MAX_OPTLEN`.
>
> Scenario shown as below:
>
>            `process A`                      `process B`
>            -----------                      ------------
>   BPF_CGROUP_GETSOCKOPT_MAX_OPTLEN
>                                             enable CGROUP_GETSOCKOPT
>   BPF_CGROUP_RUN_PROG_GETSOCKOPT (-EFAULT)
>
> To prevent this, invoke `cgroup_bpf_enabled()` only once and cache the
> result in a newly added local variable `enabled`.
> Both `BPF_CGROUP_*` macros in `do_sock_getsockopt` will then check their
> condition using the same `enabled` variable as the condition variable,
> instead of using the return values from `cgroup_bpf_enabled` called by
> themselves as the condition variable(which could yield different results).
> This ensures that either both `BPF_CGROUP_*` macros pass the condition
> or neither does.
>
> Fixes: 0d01da6afc54 ("bpf: implement getsockopt and setsockopt hooks")
> Co-developed-by: Yanghui Li <yanghui.li@...iatek.com>
> Signed-off-by: Yanghui Li <yanghui.li@...iatek.com>
> Co-developed-by: Cheng-Jui Wang <cheng-jui.wang@...iatek.com>
> Signed-off-by: Cheng-Jui Wang <cheng-jui.wang@...iatek.com>
> Signed-off-by: Tze-nan Wu <Tze-nan.Wu@...iatek.com>
> ---
>
> Chagnes from v1 to v2: https://lore.kernel.org/all/20240819082513.27176-1-Tze-nan.Wu@mediatek.com/
>   Instead of using cgroup_lock in the fastpath, invoke cgroup_bpf_enabled
>   only once and cache the value in the newly added variable `enabled`.
>   `BPF_CGROUP_*` macros in do_sock_getsockopt can then both check their
>   condition with the new variable `enable`, ensuring that either they both
>   passing the condition or both do not.
>
> Chagnes from v2 to v3: https://lore.kernel.org/all/20240819155627.1367-1-Tze-nan.Wu@mediatek.com/
>   Hide cgroup_bpf_enabled in the macro, and some modifications to adapt
>   the coding style.
>
> Chagnes from v3 to v4: https://lore.kernel.org/all/20240820092942.16654-1-Tze-nan.Wu@mediatek.com/
>   Add bpf tag to subject, and Fixes tag in body.
>
> ---
>  include/linux/bpf-cgroup.h | 15 ++++++++-------
>  net/socket.c               |  5 +++--
>  2 files changed, 11 insertions(+), 9 deletions(-)
>
> diff --git a/include/linux/bpf-cgroup.h b/include/linux/bpf-cgroup.h
> index fb3c3e7181e6..5afa2ac76aae 100644
> --- a/include/linux/bpf-cgroup.h
> +++ b/include/linux/bpf-cgroup.h
> @@ -390,20 +390,20 @@ static inline bool cgroup_bpf_sock_enabled(struct sock *sk,
>         __ret;                                                                 \
>  })
>
> -#define BPF_CGROUP_GETSOCKOPT_MAX_OPTLEN(optlen)                              \
> +#define BPF_CGROUP_GETSOCKOPT_MAX_OPTLEN(optlen, enabled)                     \
>  ({                                                                            \
>         int __ret = 0;                                                         \
> -       if (cgroup_bpf_enabled(CGROUP_GETSOCKOPT))                             \
> +       enabled = cgroup_bpf_enabled(CGROUP_GETSOCKOPT);                       \
> +       if (enabled)


I suspect the compiler generates slow code after such a patch.
pw-bot: cr

What is the problem with double cgroup_bpf_enabled() check?
yes it might return two different values, so?