Return-Path: linux-nfs-owner@vger.kernel.org Received: from fieldses.org ([174.143.236.118]:51430 "EHLO fieldses.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750901AbaJWNcf (ORCPT ); Thu, 23 Oct 2014 09:32:35 -0400 Date: Thu, 23 Oct 2014 09:32:29 -0400 To: Chuck Lever Cc: Trond Myklebust , Anna Schumaker , Linux NFS Mailing List , Tom Talpey Subject: Re: [PATCH v1 13/16] NFS: Add sidecar RPC client support Message-ID: <20141023133229.GA16717@fieldses.org> References: <20141016192919.13414.3151.stgit@manet.1015granger.net> <20141016194000.13414.83844.stgit@manet.1015granger.net> <54454762.8020506@Netapp.com> <5BF0312C-06EC-4D83-81E9-F929724A0EAD@oracle.com> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 In-Reply-To: From: "J. Bruce Fields" Sender: linux-nfs-owner@vger.kernel.org List-ID: On Tue, Oct 21, 2014 at 01:11:26PM -0400, Chuck Lever wrote: > The size of the changeset _is_ the justification. It’s > a much less invasive change to add a TCP side-car than > it is to implement RDMA backchannel on both server and > client. Something I'm confused about: is bidirectional RPC/RDMA optional or mandatory for servers to implement? Something somewhere has to be mandatory if we want to guarantee a working backchannel between any two implementations. --b.