From mboxrd@z Thu Jan 1 00:00:00 1970 From: Roland Dreier Subject: Re: [PATCH 0/21] Reliable Datagram Sockets (RDS) Date: Wed, 28 Jan 2009 14:37:25 -0800 Message-ID: References: <1233022678-9259-1-git-send-email-andy.grover@oracle.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: rds-devel@oss.oracle.com, general@lists.openfabrics.org, netdev@vger.kernel.org To: Andy Grover Return-path: Received: from sj-iport-5.cisco.com ([171.68.10.87]:5130 "EHLO sj-iport-5.cisco.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751075AbZA1Wh1 (ORCPT ); Wed, 28 Jan 2009 17:37:27 -0500 In-Reply-To: <1233022678-9259-1-git-send-email-andy.grover@oracle.com> (Andy Grover's message of "Mon, 26 Jan 2009 18:17:37 -0800") Sender: netdev-owner@vger.kernel.org List-ID: > This patchset adds support for RDS as an Infiniband ULP. RDS is an > Oracle-originated protocol used to send IPC datagrams (up to 1MB) reliably, > and is used currently in Oracle RAC and Exadata products. It's lived > in OFED for 2+ years and I think it's time to get it upstream -- most > likely into your -next tree for .30, but if it snuck into .29 via the > "new code merge-window exception" then even better. I'll read this over and comment, but to be honest I agree with Dave: this is a new socket family, and as such it belongs under net/ and probably should go through Dave's tree (just as the NFS/RDMA changes went through the NFS trees and the 9p/RDMA changes went through the 9p tree); even though it heavily uses RDMA I think the upper layer interface to sockets/networking is more relevant. - R.