On 1 November 2016 at 13:11, David Ahern <d...@cumulusnetworks.com> wrote: > On 10/31/16 6:37 PM, Thomas Graf wrote: >> - Perform simple encapsulation where offload is of no concern. >> (The existing funtionality to attach a tunnel key to the skb >> and redirect to a tunnel net_device to allow for offload >> continues to work obviously). > > have you tested the adding of headers with large packets hitting gso and > fragmentation? Wondering if mtu is used properly when headers are added and > the lwt->headroom is not accounted. See > 14972cbd34ff668c390cbd2e6497323484c9e812
Thanks for the pointer David, I will look into this. The packet size is currently limited to this when adding l2 headers: skb->dev->mtu + skb->dev->hard_header_len