Return-Path: X-Sieve: cmu-sieve 2.0 Return-Path: Received: (from majordom@localhost) by ece.cmu.edu (8.11.0/8.10.2) id h3I9Udb27878 for ips-outgoing; Fri, 18 Apr 2003 05:30:39 -0400 (EDT) X-Authentication-Warning: ece.cmu.edu: majordom set sender to owner-ips@ece.cmu.edu using -f Received: from master.linux-ide.org (astound-64-85-224-253.ca.astound.net [64.85.224.253]) by ece.cmu.edu (8.11.0/8.10.2) with ESMTP id h3I9UW327872 for ; Fri, 18 Apr 2003 05:30:36 -0400 (EDT) Received: from localhost (andre@localhost) by master.linux-ide.org (8.9.3/8.9.3) with ESMTP id CAA13705; Fri, 18 Apr 2003 02:28:14 -0700 Date: Fri, 18 Apr 2003 02:28:14 -0700 (PDT) From: Andre Hedrick To: "Sperry, Todd" cc: ips@ece.cmu.edu, "'Elizabeth G. Rodriguez'" , "'Black_David@emc.com'" Subject: Re: IPS: Adaptec IPR Notice In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Sender: owner-ips@ece.cmu.edu Precedence: bulk Basically it sounds like a simple mem-pool shared between the net-stack and storage stack. Applying ZERO-copy of d-gram from the pre-allocated pool and indexing against the offset value of a given PDU, it is simple logic to build the scatter-gather list according to the following. Command w/ or w/o Immediate data describes the initial reference point. Dump Data payload in ZeroCopy fashion to a pool of IOVECS. Carry the sum of data collected in inside your stuct index for determining all data is present. BLAH BLAH ... Once all PDU's are collected, build the SG Table for the HBA based on any given logical sorting routine. Since the IOVECS were pre allocated and of page size, the array of for the entry points and their associated offsets can be acted upon to create/produce the appearance of DataPDUInOrder=No. Now pass the simple pointer of the physical memory address containing the unclever SGT and push it to the HBA. Been doing this for quite a while and is the basics of binding a shared pool of mapped memory for a generic zero-copy between the network and spindle. Not much more than sendmail on crack with a spindle added for a sprinkle of flavor. If this is your patent, well not much there. If it is not your patent, here is a free way to bypass it. If I have to patent the method and make it free to help keep the IETF clean of patents great! This is not NCITS, and ... "reasonable, non-discriminatory terms" recall 1394 Firewire from the T commitees. In keeping with the spirit of IETF, if there are any other members who would like to derive a better mouse-trap in a joint effort to file for a patent and release it royality free for the good if iSCSI and interoperability, ping me offline. Regards, Andre Hedrick LAD Storage Consulting Group PS, I have not looked at your patent nor will I. On Thu, 17 Apr 2003, Sperry, Todd wrote: > > Adaptec has made a patent application that may relate to the > work underway in the IPS working group, specifically to the > work derived from draft-ietf-ips-iscsi-20.txt. This patent application > describes a method and process for calculating a CRC on data > received out of order without having to store and assemble the data. > > If any patents are issued to Adaptec with claims that are necessary > for practicing this standard, any party will be able to obtain the right > to implement, use and distribute the technology or works when implementing, > using or distributing technology based upon the specific specification > under openly specified, reasonable, non-discriminatory terms. > > Requests may be sent to: > > Adaptec Legal Department > 691 South Milpitas Boulevard > Milpitas, Ca. 95035 > > Phone: (408) 957-1718 > Email: legal@adaptec.com >