[Top] [All Lists]

Re: pppoe and receive checksum offload

To: "David S. Miller" <davem@xxxxxxxxxxxxx>
Subject: Re: pppoe and receive checksum offload
From: Alexey Kuznetsov <kuznet@xxxxxxxxxxxxx>
Date: Mon, 28 Feb 2005 12:21:30 +0300
Cc: Stephen Hemminger <shemminger@xxxxxxxx>, mostrows@xxxxxxxxxxxxx, kuznet@xxxxxxxxxxxxx, netdev@xxxxxxxxxxx
Domainkey-signature: a=rsa-sha1; q=dns; c=nofws; s=s1024;; b=biskZGh9mOI+TBrQzdPpH2/xgc1sYHS8mrmtyUJmS6jIOOk2zft9pMDJgshVL4RZRv1ZsfzL7dyfxird4NJJQcr5sXBJMlAV9ZIfPLqbf5JMZFMhMvTs0/XkrySck+6eMiAAtp3YtBoMikPZOTbTyUZ3C7Mf8BvQHHgUrOcUd3w=;
In-reply-to: <20050227202011.5ccefb22.davem@xxxxxxxxxxxxx>
References: <20050224155906.73890361@xxxxxxxxxxxxxxxxx> <20050227202011.5ccefb22.davem@xxxxxxxxxxxxx>
Sender: netdev-bounce@xxxxxxxxxxx
User-agent: Mutt/1.5.6i

> Changing or expanding the link level headers should only mess
> up the hw checksum if you are using CHECKSUM_HW, is that what
> your skge driver is using?

Well, even if this driver uses CHECKSUM_UNNECESSARY, the bug is still
apparently present for another devices which use CHECKSUM_HW.

Actually, the bug is in ppp_input(). It is responsible for setting correct
ip_summed before doing netif_rx(). It could even optimize subtracting
checksum of stripped ppp header from ip_summed (f.e. ipgre_rcv() do this)

But suggested fix is still correct. Unless there are another users of
ppp_input() with the same problem.


<Prev in Thread] Current Thread [Next in Thread>