netdev - Re: [PATCH] bpf: Fix unsigned 'datasec_id' compared with zero in check_pseudo_btf

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] [day] [month] [year] [list]

Message-ID: <CAEf4BzaSXd4FSM7v+HMobCSkuvATB5mZRZ6ZPzG4PzQ959jv4w@mail.gmail.com>
Date:   Tue, 10 Nov 2020 10:42:53 -0800
From:   Andrii Nakryiko <andrii.nakryiko@...il.com>
To:     Denis Kirjanov <kda@...ux-powerpc.org>
Cc:     xiakaixu1987@...il.com, Alexei Starovoitov <ast@...nel.org>,
        Daniel Borkmann <daniel@...earbox.net>,
        Martin Lau <kafai@...com>, Song Liu <songliubraving@...com>,
        Yonghong Song <yhs@...com>,
        Andrii Nakryiko <andrii@...nel.org>,
        john fastabend <john.fastabend@...il.com>,
        KP Singh <kpsingh@...omium.org>,
        Networking <netdev@...r.kernel.org>, bpf <bpf@...r.kernel.org>,
        open list <linux-kernel@...r.kernel.org>,
        Kaixu Xia <kaixuxia@...cent.com>
Subject: Re: [PATCH] bpf: Fix unsigned 'datasec_id' compared with zero in check_pseudo_btf_id

On Tue, Nov 10, 2020 at 5:02 AM Denis Kirjanov <kda@...ux-powerpc.org> wrote:
>
> On 11/10/20, xiakaixu1987@...il.com <xiakaixu1987@...il.com> wrote:
> > From: Kaixu Xia <kaixuxia@...cent.com>
> >
> > The unsigned variable datasec_id is assigned a return value from the call
> > to check_pseudo_btf_id(), which may return negative error code.
> >
> > Fixes coccicheck warning:
> >
> > ./kernel/bpf/verifier.c:9616:5-15: WARNING: Unsigned expression compared
> > with zero: datasec_id > 0
> >
> > Reported-by: Tosk Robot <tencent_os_robot@...cent.com>
> > Signed-off-by: Kaixu Xia <kaixuxia@...cent.com>
> > ---
> >  kernel/bpf/verifier.c | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
> > index 6200519582a6..e9d8d4309bb4 100644
> > --- a/kernel/bpf/verifier.c
> > +++ b/kernel/bpf/verifier.c
> > @@ -9572,7 +9572,7 @@ static int check_pseudo_btf_id(struct bpf_verifier_env
> > *env,
> >                              struct bpf_insn *insn,
> >                              struct bpf_insn_aux_data *aux)
> >  {
> > -     u32 datasec_id, type, id = insn->imm;
> > +     s32 datasec_id, type, id = insn->imm;
>
> but the value is passed as u32 to btf_type_by_id()...
>
> btf_find_by_name_kind() returns s32

Right, valid range of BTF type IDs are >= 0 and (significantly) less
than INT_MAX. So s32 is used to signal valid BTF ID or negative error,
but all the APIs accepting BTF ID accept it as just u32.

>
>
> >       const struct btf_var_secinfo *vsi;
> >       const struct btf_type *datasec;
> >       const struct btf_type *t;
> > --
> > 2.20.0
> >
> >