[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-devel] [PATCH 2/2] netfront skb padding



[...]
> Jumbo frames are a completely different ball game. Ideally you'd want to
> get the headers seperated off and have the actual data units page
> aligned.

That would require a fair amount of knowledge about the packet, which I would
argue is not the place of the virtual device driver (though it might be for 
netback).  But, better cross that bridge when we come to it.

> > Sure, but we only have 1 layer (not the 9 that we have room 
> > for).  All other 
> > layers would be encapsulated inside the data of the skb (or am I not 
> > understanding the problem correctly).
> 
> Other layers of headers and encapsulation could get added when the
> packet arrives in the backend, so having a generous headroom may be

How could their be more than one layer of encapsulation greater than MTU?  
When the packet comes into dom0, it is already <= MTU.  It gets encapsulated
with another layer of routing data, and sent to its destination.  So, it is 
at maximum size MTU+24.

> useful. Right now, I can certainly see no downside.

<Shaking fist in air> It is wasteful!!!! </shaking fist in air>  But, if it 
doesn't matter to you, then I'll move on to something else.

-- 
Jon Mason
jdmason@xxxxxxxxxx


-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxxx
https://lists.sourceforge.net/lists/listinfo/xen-devel


 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.