netdev - Re: [RFC net-next] net: guard drivers against shared skbs

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: <20211115101150.137485cb@kicinski-fedora-pc1c0hjn.dhcp.thefacebook.com>
Date:   Mon, 15 Nov 2021 10:11:50 -0800
From:   Jakub Kicinski <kuba@...nel.org>
To:     Eric Dumazet <eric.dumazet@...il.com>
Cc:     davem@...emloft.net, netdev@...r.kernel.org, hawk@...nel.org,
        syzbot+4c63f36709a642f801c5@...kaller.appspotmail.com
Subject: Re: [RFC net-next] net: guard drivers against shared skbs

On Mon, 15 Nov 2021 09:59:56 -0800 Eric Dumazet wrote:
> > The IFF_TX_SKB_SHARING flag is pretty toothless as it stands.
> 
> skb_padto() needs to be replaced by something better.
> so that skb can be cloned if needed.
> 
> 
> static inline int skb_padto(struct sk_buff *skb, unsigned int len)
> 
> ->  
> 
> static inline struct sk_buff *skb_padto(struct sk_buff *skb, unsigned int len)

Indeed, that was my first instinct but I wasn't up for fixing up all
the drivers which call skb_pad(), skb_cow_head() etc.

Let me leave this be for now..