netdev - IGB handling of zero length checksumming?

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 for Android: free password hash cracker in your pocket

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Message-Id: <20100411.024027.120459168.davem@davemloft.net>
Date:	Sun, 11 Apr 2010 02:40:27 -0700 (PDT)
From:	David Miller <davem@...emloft.net>
To:	netdev@...r.kernel.org
CC:	jeffrey.t.kirsher@...el.com, jesse.brandeburg@...el.com,
	bruce.w.allan@...el.com, alexander.h.duyck@...el.com,
	peter.p.waskiewicz.jr@...el.com, john.ronciak@...el.com
Subject: IGB handling of zero length checksumming?

If the IGB is given a "skb->ip_summed === CHECKSUM_PARTIAL" packet and
the data area past the TCP header is of zero length, will it do the
right thing?

I'm asking because I made a change recently in net-2.6 that causes TCP
packets with no data (ACKs, etc.) to start using CHECKSUM_PARTIAL.

And a user with IGB interfaces on a system running net-2.6 is getting
failures connecting to SSH and stuff like that and it is known that
this TCP commit of mine is the culprit.  See:

http://marc.info/?l=linux-netdev&m=127097744616962&w=2

I'm very likely to just revert this change regardless of what the
problem is, and try to sort this out in net-next-2.6, but it'd be nice
to know if IGB indeed has a problem in this area and that's the true
cause of the issue.

Thanks!
--
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