From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jason Gunthorpe Subject: Re: [PATCH rdma-next V1 04/10] IB/core: Support accessing SA in virtualized environment Date: Tue, 8 Mar 2016 10:54:11 -0700 Message-ID: <20160308175411.GC10805@obsidianresearch.com> References: <1457384940-11951-1-git-send-email-eli@mellanox.com> <1457384940-11951-5-git-send-email-eli@mellanox.com> <20160307220434.GB6706@obsidianresearch.com> <20160307222908.GC33629@x-vnc01.mtx.labs.mlnx> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <20160307222908.GC33629-lgQlq6cFzJSjLWYaRI30zHI+JuX82XLG@public.gmane.org> Sender: linux-rdma-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Eli Cohen Cc: liranl-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org, eitan-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org, dledford-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org, linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-Id: linux-rdma@vger.kernel.org On Tue, Mar 08, 2016 at 12:29:08AM +0200, Eli Cohen wrote: > On Mon, Mar 07, 2016 at 03:04:34PM -0700, Jason Gunthorpe wrote: > > On Mon, Mar 07, 2016 at 11:08:54PM +0200, Eli Cohen wrote: > > > > > +enum { > > > + IB_SA_WELL_KNOWN_GUID = 2, > > > +}; > > > > Where did this come from? > It comes from the virtualization annex which can be found in the ibta > web site: > SM GID - A well-known GID that is associated with the SM, comprising > the concatenation of the Subnet prefix and the GUID 0x2. The SM GID is > never present in any GID Table. If SubnetPrefix is modified by the SM > the SM GID is updated implicitly. IBTA should fix this before it gets too far. > > > > It is wrong to have a EUI-64 like this that has a 0 OUI. > > > > It is wrong to have a 'well known GUID' that is by definition not > > universal without setting the local bit (IBA spec even says this) > > > > The well known guid should still conform to IEEE rules for creating > > EUI-64s. > > > > Note, that the IBA is a little confused because it tries to follow > > IPv6 rules in some places not realizing that IPv6 uses an EUI-64 with > > the U/L bit inverted. > > > > IMHO, new constants should strictly follow the IEEE EUI-64 rules and > > not try and confuse matters futher. The unused, incorrect, IBA > > definition of ::1 should be ignored. > > > > I will let Eitan and Liran comment on this. Jason -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org More majordomo info at http://vger.kernel.org/majordomo-info.html