On Tue, Oct 27, 2015, at 17:06, Tom Herbert wrote:

On Tue, Oct 27, 2015 at 8:02 AM, Hannes Frederic Sowa
[off-list ref] wrote:

quoted

CHECKSUM_PARTIAL skbs should never arrive in ip_fragment. If we get one
of those warn about them once and handle them gracefully by recalculating
the checksum.

Cc: Eric Dumazet <edumazet@google.com>
Cc: Vlad Yasevich <redacted>
Cc: Benjamin Coddington <redacted>
Cc: Tom Herbert <redacted>
Signed-off-by: Hannes Frederic Sowa <redacted>
---
 net/ipv4/ip_output.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index 0b02417..3f94a3b 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c

@@ -533,6 +533,11 @@ int ip_do_fragment(struct net *net, struct sock *sk, struct sk_buff *skb,

        dev = rt->dst.dev;

+       /* for offloaded checksums cleanup checksum before fragmentation */
+       if (WARN_ON_ONCE(skb->ip_summed == CHECKSUM_PARTIAL) &&
+           (err = skb_checksum_help(skb)))
+               goto fail;
+

Why the WARN_ON_ONCE? Is there a prior check somewhere that avoid this
condition?

While I am pretty sure we should not hit the condition in IPv6 anymore,
I think this could frighten people in IPv4 land. I will repost without
the WARN_ON_ONCE. Maybe it makes sense to use the IFF_DEBUG interface
flags again? :)

I will repost without those WARN_ON_ONCEs.

Bye,
Hannes