From mboxrd@z Thu Jan 1 00:00:00 1970 From: Or Gerlitz Subject: Re: [PATCH RDMA] Fixup IPv6 support and IPv4 routing corner cases for RDMA CM Date: Thu, 05 Nov 2009 14:21:53 +0200 Message-ID: <4AF2C361.1080307@voltaire.com> References: <20091028054253.GA22882@obsidianresearch.com> <4E89386C612E46C9A33501CEC2521BAF@amr.corp.intel.com> <20091028165043.GD1966@obsidianresearch.com> <20091028185847.GK14520@obsidianresearch.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20091028185847.GK14520-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/@public.gmane.org> Sender: linux-rdma-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Jason Gunthorpe , Sean Hefty Cc: linux-rdma List-Id: linux-rdma@vger.kernel.org Jason Gunthorpe wrote: > Well, looking at this, I see that ADDR_CHANGE is only generated on netdev events NETDEV_BONDING_FAILOVER - so what are the semantics WRT to UCM here? As I said before, defeating the ND process seriously affects how bonding works. Again, much of the rdma-cm essence is to link/bridge/glue the netdev and rdma stacks. I don't see what it has to do with UCM which you keep mentioning/repeating over this email/thread and elsewhere. Sean, I don't see why ACM can't be implemented without introducing AF_IB/PS_IB, is it really a must or more of something you see as making the implementation somehow easier? Now, even if you are fine with AF_IB, I don't think that UCM should take over the rdma-ucm and fill the kernel rdma-cm code with tons of IB specifics code. If people need UCM, lets them use libibcm. Or. -- 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