netdev - Re: [PATCH net 2/2] sch_cake: Make sure we can write the IP header before changing DSCP bits

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, 04 Apr 2019 22:44:33 +0200
From:   Toke Høiland-Jørgensen <toke@...hat.com>
To:     Stephen Hemminger <stephen@...workplumber.org>
Cc:     David Miller <davem@...emloft.net>, netdev@...r.kernel.org,
        cake@...ts.bufferbloat.net
Subject: Re: [PATCH net 2/2] sch_cake: Make sure we can write the IP header before changing DSCP bits

Stephen Hemminger <stephen@...workplumber.org> writes:

> On Thu, 04 Apr 2019 15:01:33 +0200
> Toke Høiland-Jørgensen <toke@...hat.com> wrote:
>
>>  static u8 cake_handle_diffserv(struct sk_buff *skb, u16 wash)
>>  {
>> +	int wlen = skb_network_offset(skb);
>
> In theory this could be negative, you should handle that?
> Rather than calling may_pull() with a huge unsigned value.

Huh, that would imply that skb->network_header points to before
skb->head; when does that happen?

Also, pskb_may_pull() does check for len > skb->len, so I guess a
follow-up question would be, "does it happen often enough to warrant
handling at this level"?

Also, I copied that bit from sch_dsmark, so if you really thing it needs
to be fixed, I guess we should fix both...

-Toke