Re: [PATCH bpf-next] bpf: prevent non-IPv4 socket to be added into sock hash
From: John Fastabend <john.fastabend@gmail.com>
Date: 2018-06-01 07:56:31
From: John Fastabend <john.fastabend@gmail.com>
Date: 2018-06-01 07:56:31
On 05/31/2018 06:00 PM, Eric Dumazet wrote:
On Thu, May 31, 2018 at 7:32 PM John Fastabend [off-list ref] wrote:quoted
Hi Wei, Thanks for the report and fix. It would be better to fix the root cause so that IPv6 works as intended. I'm testing the following now, Author: John Fastabend [off-list ref] Date: Thu May 31 14:38:59 2018 -0700 sockmap: fix crash when ipv6 sock is added by adding support for IPv6 Apparently we had a testing escape and missed IPv6. This fixes a crash where we assign tcp_prot to IPv6 sockets instead of tcpv6_prot. Signed-off-by: John Fastabend [off-list ref]Hi John In any case, please forward correct attribution for Wei's work, and syzbot 'Reported-by'
Will send update with tags in a moment.
Are you sure you are handling IPv4 mapped in IPv6 sockets as well ?
No, will look into it. Although I didn't see any code to handle it in the ./net/tls case either so if there is some issue with this it could possibly exist in both ULPs. I guess if ipv4 mapped ipv6 changes prot or callbacks then we could stomp on it.
Thanks.