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: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAC5F_1mrXgEtx7sx1ihdm5EA6cxP7cA7mOjVmP8HvWxatgAcRA@mail.gmail.com>
Date:	Mon, 19 May 2014 16:05:47 -0400
From:	Michael Spang <spang@...omium.org>
To:	Eric Dumazet <eric.dumazet@...il.com>
Cc:	Patrick McHardy <kaber@...sh.net>, netdev@...r.kernel.org,
	Linux Kernel <linux-kernel@...r.kernel.org>,
	Daniel Lezcano <daniel.lezcano@...e.fr>
Subject: Re: [PATCH] macvlan: Fix checksum errors when ip_summed is CHECKSUM_PARTIAL

On Mon, May 19, 2014 at 1:44 PM, Eric Dumazet <eric.dumazet@...il.com> wrote:
>
> On Mon, 2014-05-19 at 13:24 -0400, Michael Spang wrote:
> > Changing ip_summed from CHECKSUM_PARTIAL to CHECKSUM_UNNECESSARY
> > will result in an incorrect checksum if the packet is sent off the box.
> >
> > Cc: stable@...r.kernel.org
> > Signed-off-by: Michael Spang <spang@...omium.org>
> > ---
> >  drivers/net/macvlan.c | 4 +++-
> >  1 file changed, 3 insertions(+), 1 deletion(-)
> >
> > diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c
> > index 753a8c2..806b56a 100644
> > --- a/drivers/net/macvlan.c
> > +++ b/drivers/net/macvlan.c
> > @@ -267,7 +267,9 @@ static int macvlan_queue_xmit(struct sk_buff *skb, struct net_device *dev)
> >
> >       if (vlan->mode == MACVLAN_MODE_BRIDGE) {
> >               const struct ethhdr *eth = (void *)skb->data;
> > -             skb->ip_summed = CHECKSUM_UNNECESSARY;
> > +
> > +             if (skb->ip_summed == CHECKSUM_NONE)
> > +                     skb->ip_summed = CHECKSUM_UNNECESSARY;
> >
> >               /* send to other bridge ports directly */
> >               if (is_multicast_ether_addr(eth->h_dest)) {
>
> Hi Michael
>
> Any idea why commit 12a2856b604476c27d85a5f9a57ae1661fc46019
> added this stuff then ?
>

I think that patch is wrong. It's not OK to change CHECKSUM_PARTIAL to
CHECKSUM_UNNECESSARY. This code is changing the meaning of the
ip_summed field from "the checksum is incomplete" to "the checksum is
complete and correct". So if the packet bounces around the network
stack and eventually gets sent out a physical NIC, the checksum was
never completed anywhere and the packet is dropped by the receiving
box.

Look at 0b7967503dc97864f283a3a06fbe23e041876138 for a patch that
fixes exactly this problem on veth devices.


>
> You should give more infos of your use case.
>
> CC Daniel Lezcano <daniel.lezcano@...e.fr>
>
> I removed stable@ , as stable submissions are not done like that.
>

Perhaps I've misunderstood something, but this suggestion is right
from https://www.kernel.org/doc/Documentation/stable_kernel_rules.txt

 - To have the patch automatically included in the stable tree, add the tag
     Cc: stable@...r.kernel.org
   in the sign-off area. Once the patch is merged it will be applied to
   the stable tree without anything else needing to be done by the author
   or subsystem maintainer.
--
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

Powered by Openwall GNU/*/Linux Powered by OpenVZ