Re: [Linux-kernel-mentees] [PATCH] bpf: Fix NULL pointer dereference in __btf_resolve_helper_id()

From: Andrii Nakryiko
Date: Mon Jul 13 2020 - 20:53:45 EST


On Mon, Jul 13, 2020 at 5:43 PM Peilin Ye <yepeilin.cs@xxxxxxxxx> wrote:
>
> Prevent __btf_resolve_helper_id() from dereferencing `btf_vmlinux`
> as NULL. This patch fixes the following syzbot bug:
>
> https://syzkaller.appspot.com/bug?id=5edd146856fd513747c1992442732e5a0e9ba355
>
> Reported-by: syzbot+ee09bda7017345f1fbe6@xxxxxxxxxxxxxxxxxxxxxxxxx
> Signed-off-by: Peilin Ye <yepeilin.cs@xxxxxxxxx>
> ---
> kernel/bpf/btf.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c
> index 30721f2c2d10..3e981b183fa4 100644
> --- a/kernel/bpf/btf.c
> +++ b/kernel/bpf/btf.c
> @@ -4088,7 +4088,7 @@ static int __btf_resolve_helper_id(struct bpf_verifier_log *log, void *fn,
> const char *tname, *sym;
> u32 btf_id, i;
>
> - if (IS_ERR(btf_vmlinux)) {
> + if (IS_ERR_OR_NULL(btf_vmlinux)) {
> bpf_log(log, "btf_vmlinux is malformed\n");

Can you please split IS_ERR and NULL cases and emit different messages
to log? If the kernel is not built with btf_vmlinux, saying that it's
"malformed" will confuse people. Thanks.

> return -EINVAL;
> }
> --
> 2.25.1
>