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]
Date:   Fri, 23 Apr 2021 13:44:40 -0700 (PDT)
From:   David Miller <davem@...emloft.net>
To:     sd@...asysnail.net
Cc:     phil@...lpotter.co.uk, kuba@...nel.org, netdev@...r.kernel.org,
        linux-kernel@...r.kernel.org, edumazet@...gle.com, fw@...len.de
Subject: Re: [PATCH v2] net: geneve: modify IP header check in
 geneve6_xmit_skb and geneve_xmit_skb

From: Sabrina Dubroca <sd@...asysnail.net>
Date: Fri, 23 Apr 2021 15:41:17 +0200

> 2021-04-23, 00:49:45 +0100, Phillip Potter wrote:
>> Modify the header size check in geneve6_xmit_skb and geneve_xmit_skb
>> to use pskb_inet_may_pull rather than pskb_network_may_pull. This fixes
>> two kernel selftest failures introduced by the commit introducing the
>> checks:
>> IPv4 over geneve6: PMTU exceptions
>> IPv4 over geneve6: PMTU exceptions - nexthop objects
>> 
>> It does this by correctly accounting for the fact that IPv4 packets may
>> transit over geneve IPv6 tunnels (and vice versa), and still fixes the
>> uninit-value bug fixed by the original commit.
>> 
>> Reported-by: kernel test robot <oliver.sang@...el.com>
>> Fixes: 6628ddfec758 ("net: geneve: check skb is large enough for IPv4/IPv6 header")
>> Suggested-by: Sabrina Dubroca <sd@...asysnail.net>
>> Signed-off-by: Phillip Potter <phil@...lpotter.co.uk>
>> ---
>> 
>> V2:
>> * Incorporated feedback from Sabrina Dubroca regarding pskb_inet_may_pull
>> * Added Fixes: tag as requested by Eric Dumazet
> 
> Thanks Phillip.
> 
> Acked-by: Sabrina Dubroca <sd@...asysnail.net>
> 
> Jakub/David, it would be great if this could get in 5.12, otherwise
> geneve is a bit broken:
> https://bugzilla.kernel.org/show_bug.cgi?id=212749

I don't think we will submit another pull req for networking, but thatys
ok it'll end up in stable eventually.

Thanks.

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ