netdev - Re: Confused about ip_summed member in sk

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 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:   Sat, 22 Oct 2022 12:51:26 -0700
From:   Cong Wang <xiyou.wangcong@...il.com>
To:     "J.J. Mars" <mars14850@...il.com>
Cc:     netdev@...r.kernel.org
Subject: Re: Confused about ip_summed member in sk_buff

On Fri, Oct 21, 2022 at 02:29:26PM +0800, J.J. Mars wrote:
> Hi everyone, I'm new here and I hope this mail won't disturb you :)
> 
> Recently I was working with something about ip_summed, and I'm really
> confused about the question what does ip_summed exactly mean?
> This member is defined with comment Driver fed us an IP checksum'. So
> I guess it's about IP/L3 checksum status.
> But the possible value of ip_summed like CHECKSUM_UNNECESSARY is about L4.
> 
> What confused me a lot is ip_summed seems to tell us the checksum of
> IP/L3 layer is available from its name.
> But it seems to tell us the checksum status of L4 layer from its value.
> 
> Besides, in ip_rcv() it seems the ip_summed is not used before
> calculating the checksum of IP header.
> 
> So does ip_summed indicate the status of L3 checksum status or L4
> checksum status?
> If L4, why is it named like that?

The name itself is indeed confusing, however, there are some good
explanations in the code, at the beginning of include/linux/skbuff.h. I
think that could help you to clear your confusions here.

Thanks.