Return-Path: Received: from userp1040.oracle.com ([156.151.31.81]:33690 "EHLO userp1040.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751152AbdJ0SGL (ORCPT ); Fri, 27 Oct 2017 14:06:11 -0400 Content-Type: text/plain; charset=us-ascii Mime-Version: 1.0 (Mac OS X Mail 9.3 \(3124\)) Subject: Re: Draft RFC for ONC RPC over AF_VSOCK From: Chuck Lever In-Reply-To: <1509127168.4946.14.camel@redhat.com> Date: Fri, 27 Oct 2017 14:06:00 -0400 Cc: Matt Benjamin , Stefan Hajnoczi , Linux NFS Mailing List , Bruce Fields , Steve Dickson , Anna Schumaker , Trond Myklebust , Daniel Berrange , NFSv4 Message-Id: References: <20171005200835.GA31525@stefanha-x1.localdomain> <1509110202.4704.7.camel@redhat.com> <1509127168.4946.14.camel@redhat.com> To: Jeff Layton Sender: linux-nfs-owner@vger.kernel.org List-ID: > On Oct 27, 2017, at 1:59 PM, Jeff Layton wrote: > > I agree -- that could be useful later. Given that, maybe we should call > the netids something like: > > vsockc: connected vsock > vsockd: datagram vsock > > AIUI, netids are just something we inherited from Sun when we got the > TI-RPC library. I don't think they are governed by any sort of > names+numbers authority, are they? Jeff, the relevant authority is IANA, and that's the whole point of this I-D: to request netid assignments and specify the universal address format for the VSOCK AF. > If not then we're probably define it to whatever we wish, though it > might be a good idea to talk to the Solaris folks and see if they have > any input as to the naming. > > -- Jeff > > On Fri, 2017-10-27 at 09:27 -0400, Matt Benjamin wrote: >> Hi Jeff, >> >> This doc says they are: >> https://vmsplice.net/~stefan/stefanha-kvm-forum-2015.pdf >> >> But only stream sockets are mentioned here: >> https://wiki.qemu.org/Features/VirtioVsock >> >> Trond and Chuck suggested in an offline conversation a few weeks ago >> that they could imagine a datagram version of the transport being >> useful. It's probably worth passing that alone. >> >> Matt >> >> On Fri, Oct 27, 2017 at 9:16 AM, Jeff Layton wrote: >>> On Thu, 2017-10-05 at 16:50 -0400, Matt Benjamin wrote: >>>> Hi Stefan, >>>> >>>> On Thu, Oct 5, 2017 at 4:08 PM, Stefan Hajnoczi wrote: >>>>> I have previously submitted patches that implement NFS client and nfsd >>>>> support for the AF_VSOCK address family. In order for this to be >>>>> acceptable for merge the AF_VSOCK transport needs to be defined in an >>>>> IETF RFC. Below is a draft RFC that defines ONC RPC over AF_VSOCK. >>>>> >>>>> My patches use netid "vsock" but "tcpv" has also been suggested. This draft >>>>> RFC still uses "vsock" but I'll update it to "tcpv" if there is consensus. >>>>> >>>> >>>> I think "vsock" is the appropriate netid, not "tcpv." Stream >>>> orientation, if anything, is the general category containing TCP and >>>> VSOCK, not the reverse. But really I think it's just more clear. >>>> >>> >>> Agreed. VSOCK is its own thing. It bears some resemblance to TCP, but >>> calling it tcpv would be confusing. IIRC, Chuck only proposed that when >>> we were discussing an alternative transport that would look more like a >>> typical network. >>> >>> BTW: Does VSOCK have a connectionless mode, analogous to UDP? If so, >>> then it may be nice to consider what the netid for that might look like >>> as well, before we settle on any names. >>> -- >>> Jeff Layton >> >> >> > > -- > Jeff Layton -- Chuck Lever