[<prev] [next>] [day] [month] [year] [list]
Message-ID: <CA+mtBx-WURWvoj9CHRD_-_Ko6k3D-DJbJCqtcUao72jqSZxKVQ@mail.gmail.com>
Date: Wed, 21 May 2014 14:15:33 -0700
From: Tom Herbert <therbert@...gle.com>
To: David Miller <davem@...emloft.net>
Cc: Linux Netdev List <netdev@...r.kernel.org>
Subject: Re: [PATCH 1/5 v2] sunrpc: Remove sk_no_check setting
On Wed, May 21, 2014 at 2:14 PM, Tom Herbert <therbert@...gle.com> wrote:
>
>
>
>
> On Wed, May 21, 2014 at 1:37 PM, David Miller <davem@...emloft.net> wrote:
>>
>> From: Tom Herbert <therbert@...gle.com>
>> Date: Mon, 19 May 2014 10:09:56 -0700 (PDT)
>>
>> > Setting sk_no_check to UDP_CSUM_NORCV seems to have no effect.
>> >
>> > Signed-off-by: Tom Herbert <therbert@...gle.com>
>>
>> It does have an effect.
>>
>> All of the UDP receive paths check for sk->sk_no_check being non-zero,
>> causing it to accept packets with a zero checksum.
>>
>> You're changing behavior here.
>
>
> We always accept zero UDP checksums for IPv4, there is currently no way to disable that (sk_no_check should not affect that). For IPv6 UDP, I intend to make allowing zero RX checksums not to be default and must be explicitly configured. I'm not sure if sunrpc qualifies as a protocol for which zero v6 csums should be allowed, but in any case since we weren't accepting any v6 zero csums before there is no behavior change in that case either.
>
--
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
Powered by blists - more mailing lists