Peter Memishian wrote:

It seems that all that does today is checksums, but that they *may* throw
a bunch of other stuff in there in the future.  I'm not sure I like the
idea of defining that nebulous a function.

In any case, it seems like this would be worthwhile to work through with
Nicolas Droux and the Nemo team.

I think it's an interesting concept, and I can certainly see the benefits for offloading capabilities.

The Nemo (aka GLDv3) interfaces allows us to clearly identify the various capabilities supported, and we have an opportunity to be more explicit about the processing requested by the driver.

If someone is interested to work on this directly, let me know. It's certainly an effort we can pursue under the OpenSolaris Nemo project, see http://opensolaris.org/os/project/nemo

Nicolas.

--
Nicolas Droux, Solaris Kernel Networking
Sun Microsystems, Inc. http://blogs.sun.com/droux

_______________________________________________
networking-discuss mailing list
[email protected]

Reply via email to