netdev - Re: Accessing XDP packet memory from the end

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]

Date:   Thu, 21 Apr 2022 18:27:47 +0200
From:   Jesper Dangaard Brouer <jbrouer@...hat.com>
To:     Larysa Zaremba <larysa.zaremba@...el.com>,
        bpf <bpf@...r.kernel.org>
Cc:     brouer@...hat.com, netdev <netdev@...r.kernel.org>,
        Andrii Nakryiko <andrii@...nel.org>,
        Alexei Starovoitov <ast@...nel.org>,
        Daniel Borkmann <daniel@...earbox.net>,
        Toke Hoiland-Jorgensen <toke@...hat.com>,
        Magnus Karlsson <magnus.karlsson@...el.com>,
        Maciej Fijalkowski <maciej.fijalkowski@...el.com>,
        Alexander Lobakin <alexandr.lobakin@...el.com>,
        "xdp-hints@...-project.net" <xdp-hints@...-project.net>
Subject: Re: Accessing XDP packet memory from the end



On 21/04/2022 17.56, Larysa Zaremba wrote:
> Dear all,
> Our team has encountered a need of accessing data_meta in a following way:
> 
> int xdp_meta_prog(struct xdp_md *ctx)
> {
> 	void *data_meta_ptr = (void *)(long)ctx->data_meta;
> 	void *data_end = (void *)(long)ctx->data_end;
> 	void *data = (void *)(long)ctx->data;
> 	u64 data_size = sizeof(u32);
> 	u32 magic_meta;
> 	u8 offset;
> 
> 	offset = (u8)((s64)data - (s64)data_meta_ptr);

I'm not sure the verifier can handle this 'offset' calc. As it cannot
statically know the sized based on this statement. Maybe this is not the
issue.

> 	if (offset < data_size) {
> 		bpf_printk("invalid offset: %ld\n", offset);
> 		return XDP_DROP;
> 	}
> 
> 	data_meta_ptr += offset;
> 	data_meta_ptr -= data_size;
> 
> 	if (data_meta_ptr + data_size > data) {
> 		return XDP_DROP;
> 	}
> 		
> 	magic_meta = *((u32 *)data);
> 	bpf_printk("Magic: %d\n", magic_meta);
> 	return XDP_PASS;
> }
> 
> Unfortunately, verifier claims this code attempts to access packet with
> an offset of -2 (a constant part) and negative offset is generally forbidden.

Are you forgetting to mention:
  - Have you modified the NIC driver to adjust data_meta pointer and 
provide info in this area?

p.s. this is exactly what I'm also working towards[1], so I'll be happy
to collaborate.  I'm missing the driver code, as link[1] is focused on
decoding BTF data_meta area in userspace for AF_XDP.

[1] 
https://github.com/xdp-project/bpf-examples/tree/master/AF_XDP-interaction

> For now we have 2 solutions, one is using bpf_xdp_adjust_meta(),
> which is pretty good, but not ideal for the hot path.
> The second one is the patch at the end.
> 

Are you saying, verifier cannot handle that driver changed data_meta 
pointer and provided info there (without calling bpf_xdp_adjust_meta)?


> Do you see any other way of accessing memory from the end of data_meta/data?
> What do you think about both suggested solutions?
> 
> Best regards,
> Larysa Zaremba
> 
> ---
> 
> --- a/kernel/bpf/verifier.c
> +++ b/kernel/bpf/verifier.c
> @@ -3576,8 +3576,11 @@ static int check_packet_access(struct bpf_verifier_env *env, u32 regno, int off,
>   	}
>   
>   	err = reg->range < 0 ? -EINVAL :
> -	      __check_mem_access(env, regno, off, size, reg->range,
> -				 zero_size_allowed);
> +	      __check_mem_access(env, regno, off + reg->smin_value, size,
> +				 reg->range + reg->smin_value, zero_size_allowed);
> +	err = err ? :
> +	      __check_mem_access(env, regno, off + reg->umax_value, size,
> +				 reg->range + reg->umax_value, zero_size_allowed);
>   	if (err) {
>   		verbose(env, "R%d offset is outside of the packet\n", regno);
>   		return err;
>