Jason Gunthorpe wrote:
On Wed, Sep 05, 2007 at 11:35:06PM +0300, Or Gerlitz wrote:
I guess by "RC" you mean connected mode. The connected mode is now
implemented over RC but as was discussed over this list few times, it
should (and it would) move to use UC, which is also much easier to
implement in hw based gateways. Anyway, your idea to allow this
feature coming into play only under negotiation schem sounds fine to
me, however:
Sure.. Though, I'm not sure what advantage UC/RC brings to a gateway app
when you can't pass 64k MTU onto ethernet...
When the Ethernet side supports 9K Jumbo frames, if connected mode comes
into play then there should be a performance increase, so the gateway
negotiates the MTU to be 9K and so on.
Right, but I'm not suggesting using the chips offload.
Micheal has made it so you can use 'csum offload' (via disabling csum)
on any nic. You can also do the same kind of thing for TSO/GSO. If you
send jumbo TSO/GSO packets in a chunk the receiver can then do
LRO. Win all around. Sort of like jumbo MTU but without actually
changing the MTU.
This is all basically the same set of techniques we see between a
Linux guest and the linux host in a virtualization environment.
Thanks for the clarification, I have to do some catchup here on the
details re TSO/GSO and their relation to virtualization, however, to
make things a little clearer to me, do you agree that as James pointed
over this thread in
A (IB) ---- B (Gateway eg HW based) ---- C (Ethernet)
scheme, in case A does not compute the TCP checksum of a packet, its
note the role of the gateway to do so, and C would just drop it?!
Or.
_______________________________________________
general mailing list
[email protected]
http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general
To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
