linux-hardening - Re: [PATCH 1/2] wifi: ath11k: annotate skb of struct ath11k_ce_ring with __counted

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: <be969496-0c74-400c-b856-9524746a5ee2@quicinc.com>
Date: Mon, 21 Oct 2024 09:30:30 -0700
From: Jeff Johnson <quic_jjohnson@...cinc.com>
To: Dmitry Antipov <dmantipov@...dex.ru>, Jeff Johnson <jjohnson@...nel.org>
CC: Kalle Valo <kvalo@...nel.org>, Kees Cook <kees@...nel.org>,
        <linux-hardening@...r.kernel.org>, <linux-wireless@...r.kernel.org>,
        <lvc-project@...uxtesting.org>
Subject: Re: [PATCH 1/2] wifi: ath11k: annotate skb of struct ath11k_ce_ring
 with __counted_by

On 10/21/2024 7:27 AM, Dmitry Antipov wrote:
> According to 'ath11k_ce_alloc_ring()', annotate flexible array
> member 'skb' of 'struct ath11k_ce_ring' with '__counted_by()'
> to improve runtime bounds checking when CONFIG_UBSAN_BOUNDS is
> enabled. Compile tested only.
> 
> Signed-off-by: Dmitry Antipov <dmantipov@...dex.ru>
> ---
>  drivers/net/wireless/ath/ath11k/ce.h | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/net/wireless/ath/ath11k/ce.h b/drivers/net/wireless/ath/ath11k/ce.h
> index bcde2fcf02cf..9c54244970ce 100644
> --- a/drivers/net/wireless/ath/ath11k/ce.h
> +++ b/drivers/net/wireless/ath/ath11k/ce.h
> @@ -162,7 +162,7 @@ struct ath11k_ce_ring {
>  	u32 hal_ring_id;
>  
>  	/* keep last */
> -	struct sk_buff *skb[];
> +	struct sk_buff *skb[] __counted_by(nentries);
>  };
>  
>  struct ath11k_ce_pipe {

I won't be taking any of these without testing. I don't want to introduce
warnings due to the array being populated before setting the counter.

At a minimum you need to document where the counter is being set and verify
that is always done before accessing the flexible array.

(If it isn't broken, don't fix it)

/jeff