From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jason Gunthorpe Subject: Re: [PATCH for-next V5 00/12] Move RoCE GID management to IB/Core Date: Wed, 10 Jun 2015 22:07:17 -0600 Message-ID: <20150611040717.GB16599@obsidianresearch.com> References: <1433772735-22416-1-git-send-email-matanb@mellanox.com> <1828884A29C6694DAF28B7E6B8A82373A8FE5D17@ORSMSX109.amr.corp.intel.com> <55769561.8000300@mellanox.com> <5577FAFB.8020205@mellanox.com> <20150610150010.GA11243@obsidianresearch.com> <1433981756.71666.60.camel@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <1433981756.71666.60.camel-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org> Sender: linux-rdma-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: Doug Ledford Cc: Or Gerlitz , "Hefty, Sean" , Matan Barak , Moni Shoua , Somnath Kotur , "linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org" List-Id: linux-rdma@vger.kernel.org On Wed, Jun 10, 2015 at 08:15:56PM -0400, Doug Ledford wrote: > I'm not sure the complexity here is "latent RoCEv2" stuff versus simple > over-design. Well, for instance, the wrong RCU locking around table->data_vec[ix].attr.ndev appears to exist to support find_gid when called with GID_ATTR_FIND_MASK_NETDEV outside callers that hold table->lock. However, the unlocked call pattern is never used with GID_ATTR_FIND_MASK_NETDEV - that possibility was punted in v5, but the code to support it and the broken RCU to help it are still present, and will eventually be needed by rocev2 and roce namespaces.. That is just one small example, I fully expect there are others, and I think your remark about 'over-design' is likely equally contributing to the size expansion as well. 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