Thanks. I will read that and keep the arguments in mind. However, at this point this is a customer driven request. If a customer comes to us and says "I can get full offload from Vendor X... can you guys do it? We'll buy your product if you can." Then powers above me will make the choice to implement or not based on $$ and $$ alone. I might be the one to implement, but I don't make the decision whether or not to implement. -----Original Message----- From: David S. Miller [mailto:davem@redhat.com] Sent: Thursday, June 05, 2003 12:13 AM To: Jamie.Esliger@SiliconAccess.com Cc: linux-net@vger.kernel.org Subject: Re: Full TCP Offload? From: Jamie Esliger <Jamie.Esliger@SiliconAccess.com> Date: Wed, 4 Jun 2003 13:54:01 -0400 Hi guys. I'm investigating the implementation of TCP Offload features in our Network Processor. Please read this paper before making this grave mistake: http://www.usenix.org/events/hotos03/tech/full_papers/mogul/mogul.pdf I agree with the author up until he starts talking about RDMA, RDMA is not the solution as much as TOE is not the solution. - : send the line "unsubscribe linux-net" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
