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

Re: [Xen-devel] [0/5] [NET]: Add TSO support

On 27 Jun 2006, at 13:02, Herbert Xu wrote:

The following patches add TCP Segmentation Offload (TSO) support in the
domU => dom0 direction. If everyone's happy with this approach then it's
trivial to do the same thing for the opposite direction.

I've checked in all but the netfront patch. The glitches so far are:

1. The GSO patch broke netback, because netback/interface.c accesses dev->xmit_lock. None of your patches fixed this and you can't build the driver unless it's fixed -- did you somehow miss that file from one of the patches you sent?

2. The new 'wire format' with netif_tx_extra: I placed the GSO fields inside a struct inside a union, so we can extend the union with other extra-info types in future. I hope that's okay and in line with what you intended.

3. Wire format again: we need some extra documentation and info in netif.h for the new GSO fields. Currently they conveniently directly correspond to fields in a Linux skbuff: you read them out in netfront and write them straight back in netback. That's fine for Linux for now, but not so good for other OSes, nor potentially if the Linux GSO internals change later.

In particular the gso_type field is concerning. We should provide defines for the legitimate values of that field in netif.h, with a comment explaining what each one means. Extra comments are also required for the other two fields, to convince us that they aren't Linux-specific in some way. Some brief info about how GSO works in general, including usage of those fields, would help.

This is why I haven't added the netfront patch yet -- I don't want domU's using the new interface until we're satisfied we're not going to have to change the interface and break compatibility.


Xen-devel mailing list



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