From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932477AbWGHBtb (ORCPT ); Fri, 7 Jul 2006 21:49:31 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1751298AbWGHBtb (ORCPT ); Fri, 7 Jul 2006 21:49:31 -0400 Received: from rwcrmhc14.comcast.net ([216.148.227.154]:39408 "EHLO rwcrmhc14.comcast.net") by vger.kernel.org with ESMTP id S1751297AbWGHBta (ORCPT ); Fri, 7 Jul 2006 21:49:30 -0400 Message-ID: <001001c6a231$cef0c2a0$0201a8c0@OFFICEPC> From: "Accu-Tech" To: Subject: lock-free rcu-based ref-counting patent applications... Date: Fri, 7 Jul 2006 18:57:03 -0700 MIME-Version: 1.0 Content-Type: text/plain; format=flowed; charset="iso-8859-1"; reply-type=original Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2900.2869 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2900.2869 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Does anybody know if there are any patent applications for any of the "existing" RCU-based reference counting techniques? In particular, any counting algorithms that have fine granularity and do not use any atomic-ops and/or membars... The existing RCU counting implementations that I have seen use a per-object counter adjusted with CAS/membar, or actually defer count adjustments until after sync epoch (rcu-grace)... Pretty expensive imho... The reason I ask is because I have a patent application out on another technique that gets around SMR's atomic-ops/membars and most of the caveats that come along with RCU read-side "critical-regions"... I was just wondering if anybody might be working on something similar... I remember scouring through RCU/SMR-based ref-counting patents and applications during the patent-feasibility study I did a couple of years ago... I did find anything that was similar to my technique...