Anton Blanchard <anton@xxxxxxxxx> wrote:
>
> 1. tcpdump doesnt report local TSO packets, it simply prints bad-len. I
> think this is because the e1000 driver is zeroing out the IP header
> length in e1000_tso:
>
> skb->nh.iph->tot_len = 0;
>
> Does the card require this for TSO to operate? Ive worked around it in
> tcpdump for the time being.
This is a bug in e1000. Even if it is required it isn't allowed to
modify a cloned packet. It'll need to copy it so that other clone
users aren't affected.
> Now if we do a single direction test it takes forever for TSO to kick
> in. At the moment Im not sure why this is happening, tcp_current_mss
> seems to grow correctly in both cases. Is there somewhere else we are
> capping TSO packets?
What call does your application use to do the sending?
Cheers,
--
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <herbert@xxxxxxxxxxxxxxxxxxx>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
|