netdev - Re: [PATCH] cls_u32: use skb_copy_bits() to dereference data safely

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: <20100602.054520.228955151.davem@davemloft.net>
Date:	Wed, 02 Jun 2010 05:45:20 -0700 (PDT)
From:	David Miller <davem@...emloft.net>
To:	hadi@...erus.ca
Cc:	xiaosuo@...il.com, netdev@...r.kernel.org
Subject: Re: [PATCH] cls_u32: use skb_copy_bits() to dereference data safely

From: jamal <hadi@...erus.ca>
Date: Wed, 02 Jun 2010 08:25:38 -0400

> --- a/net/sched/cls_u32.c
> +++ b/net/sched/cls_u32.c
> @@ -135,6 +135,9 @@ next_knode:
>  
>  for (i = n->sel.nkeys; i>0; i--, key++) {
>  
> +        int toff = key->off+(off2&key->offmask)- 4;
> +        if (unlikely(toff > skb->len))
> +              /* bailout here - needs some thought */
>          if ((*(__be32*)(ptr+key->off+(off2&key->offmask))^key->v

I don't think it's that simple.

You can't dereference from the skb->data linear area if your offset is
beyond "skb->len - skb->data_len" (aka. skb_headlen()) since that's
where the paged or fragmented portion starts.

We really need to use skb_copy_bits() if we want to allow
any offset into the SKB, and because of all the ways
packets can be transformed and constructed we absolutely
have to.

--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html