On Fri, Sep 23, 2016 at 4:41 PM, Sowmini Varadhan <sowmini.varad...@oracle.com> wrote: > On (09/23/16 10:38), Alexander Duyck wrote: >> >> So basically what I was thinking is we do something like reserving >> NET_IP_ALIGN and continue writing headers to skb->data, but we force >> the tracking for the inner headers into frag[0] so that we can keep >> the inner headers aligned without messing up the alignment for outer >> headers. In theory the inner offset and all that would still be >> functional but might need a few tweaks. You could probably even use >> the skb->encapsulation bit to indicate you are doing this. You could >> almost think of it as us doing something like the inverse of >> pskb_pull_tail. The general idea here is we want to actually leave >> the data in skb->data, but just reference it from frag[0] so that we >> don't accidentally pull in the 2 byte padding for alignment when >> transmitting the frame. > > yes, I think something along this line could do the trick.. I tried > hacking it a bit today for vxlan, and it could be extended for all > these encaps protocols. Let me fix/test this more next week, maybe > we can discuss in Tokyo.
Agreed. Keep in mind we only really need it for the architectures that need to set NET_IP_ALIGN so we may want to end up wrapping the code in ifndef checks for HAVE_EFFICIENT_UNALIGNED_ACCESS. - Alex