From: Trond Myklebust Subject: Re: [NLM] 2.6.27.14 breakage when grace period expires Date: Thu, 12 Feb 2009 17:19:18 -0500 Message-ID: <1234477158.7190.198.camel@heimdal.trondhjem.org> References: <20090211112318.GA29133@janus> <20090211203555.GC27686@fieldses.org> <20090211203703.GA9662@janus> <20090211203948.GD27686@fieldses.org> <20090212142830.GA28107@janus> <1234451789.7190.38.camel@heimdal.trondhjem.org> <20090212153634.GB28107@janus> <1234462647.7190.53.camel@heimdal.trondhjem.org> <20090212182943.GA1945@janus> <1234465837.7190.62.camel@heimdal.trondhjem.org> <03CEE6BB-D39D-47A5-BB6C-2FB3E206142F@oracle.com> <1234467795.7190.70.camel@heimdal.trondhjem.org> <1234470457.7190.106.camel@heimdal.trondhjem.org> <2171ACCA-5CF5-425D-B7A4-CFB17DA61F75@oracle.com> <1234476134.7190.187.camel@heimdal.trondhjem.org> <446F0762-CBDF-4DF6-A510-8B56EC9ABD9D@oracle.com> Mime-Version: 1.0 Content-Type: text/plain Cc: Frank van Maarseveen , "J. Bruce Fields" , Linux NFS mailing list To: Chuck Lever Return-path: Received: from mail-out2.uio.no ([129.240.10.58]:53197 "EHLO mail-out2.uio.no" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1761293AbZBLWTZ (ORCPT ); Thu, 12 Feb 2009 17:19:25 -0500 In-Reply-To: <446F0762-CBDF-4DF6-A510-8B56EC9ABD9D@oracle.com> Sender: linux-nfs-owner@vger.kernel.org List-ID: On Thu, 2009-02-12 at 17:11 -0500, Chuck Lever wrote: > On Feb 12, 2009, at 5:02 PM, Trond Myklebust wrote: > > On Thu, 2009-02-12 at 15:43 -0500, Chuck Lever wrote: > >> The reason we might care in this case is nlm_cmp_addr() is executed > >> more frequently than nfs_sockaddr_match_ipaddr(). > > > > Actually, I'm not sure this assertion is correct. The only users of > > nlm_cmp_addr() are nlmclnt_grant(), nlm_lookup_host() and > > nlmsvc_unlock_all_by_ip(). > > > > AFAICS, the only one that needs to be v4 mapped should be > > nlmclnt_grant, > > which is not in a performance critical path... > > So then your proposal is to ensure the two arguments of the > nlm_cmp_addr() callsite in nlmclnt_grant() are both AF_INET6? Yup... I can't see that the other two callsites need anything like that.