On 5/13/25 9:51 AM, Antonio Quartulli wrote:

On 13/05/2025 09:37, Paolo Abeni wrote:

quoted

On 5/9/25 4:26 PM, Antonio Quartulli wrote:

quoted

IPv6 user packets (sent over the tunnel) may be larger than
the outgoing interface MTU after encapsulation.
When this happens ovpn should allow the kernel to fragment
them because they are "locally generated".

To achieve the above, we must set skb->ignore_df = 1
so that ip6_fragment() can be made aware of this decision.

Why the above applies only to IPv6? AFAICS the same could happen even
for IPv4.

For IPv4 we have the 'df=0' param that is passed to 
udp_tunnel_xmit_skb(), which basically leads to the same result.

You need to include (an expanded/more describing version of) the above
in the commit message.

/P