linux-kernel - Re: [PATCH] bpf: restrict verifier access to bpf_lru

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: <CAADnVQJ6_pB8ZU2Cw5S6nB4J-6s7bw5Fp-Hst9M_EE9=HxN8+g@mail.gmail.com>
Date: Tue, 15 Jul 2025 07:49:33 -0700
From: Alexei Starovoitov <alexei.starovoitov@...il.com>
To: Shankari Anand <shankari.ak0208@...il.com>, Martin KaFai Lau <martin.lau@...nel.org>
Cc: bpf <bpf@...r.kernel.org>, LKML <linux-kernel@...r.kernel.org>, 
	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>, 
	Stanislav Fomichev <sdf@...ichev.me>, Hao Luo <haoluo@...gle.com>, Jiri Olsa <jolsa@...nel.org>, 
	syzbot+ad4661d6ca888ce7fe11@...kaller.appspotmail.com
Subject: Re: [PATCH] bpf: restrict verifier access to bpf_lru_node.ref

On Tue, Jul 15, 2025 at 12:58 AM Shankari Anand
<shankari.ak0208@...il.com> wrote:
>
> syzbot reported a data race on the `ref` field of `struct bpf_lru_node`:
> https://syzkaller.appspot.com/bug?extid=ad4661d6ca888ce7fe11
>
> This race arises when user programs read the `.ref` field from a BPF map
> that uses LRU logic, potentially exposing unprotected state.
>
> Accesses to `ref` are already wrapped with READ_ONCE() and WRITE_ONCE().
> However, the BPF verifier currently allows unprivileged programs to
> read this field via BTF-enabled pointer, bypassing internal assumptions.
>
> To mitigate this, the verifier is updated to disallow access
> to the `.ref` field in `struct bpf_lru_node`.
> This is done by checking both the base type and field name
> in `check_ptr_to_btf_access()` and returning -EACCES if matched.
>
> Reported-by: syzbot+ad4661d6ca888ce7fe11@...kaller.appspotmail.com
> Closes: https://lore.kernel.org/all/6847e661.a70a0220.27c366.005d.GAE@google.com/T/
> Signed-off-by: Shankari Anand <shankari.ak0208@...il.com>
> ---
>  kernel/bpf/verifier.c | 13 +++++++++++++
>  1 file changed, 13 insertions(+)
>
> diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
> index 169845710c7e..775ce454268c 100644
> --- a/kernel/bpf/verifier.c
> +++ b/kernel/bpf/verifier.c
> @@ -7159,6 +7159,19 @@ static int check_ptr_to_btf_access(struct bpf_verifier_env *env,
>                 }
>
>                 ret = btf_struct_access(&env->log, reg, off, size, atype, &btf_id, &flag, &field_name);
> +
> +               /* Block access to sensitive kernel-internal fields */

This makes no sense. Tracing bpf progs are allowed to read
all kernel internal data fields.

Also you misread the kcsan report.

It says that 'read' comes from:

read to 0xffff888118f3d568 of 4 bytes by task 4719 on cpu 1:
 lookup_nulls_elem_raw kernel/bpf/hashtab.c:643 [inline]

which is reading hash and key of htab_elem while
write side actually writes hash too:
*(u32 *)((void *)node + lru->hash_offset) = hash;

Martin,
is it really possible for these read/write to race ?

--
pw-bot: cr