From mboxrd@z Thu Jan 1 00:00:00 1970 From: Peter Staubach Subject: Re: [PATCH 002 of 2] Avoid flush-when-locking when interclient consistency not needed. Date: Mon, 16 Apr 2007 10:48:31 -0400 Message-ID: <46238CBF.4090700@redhat.com> References: <20070306163710.8128.patches@notabene> <1070306054029.8360@suse.de> <1173192029.6393.52.camel@heimdal.trondhjem.org> <17903.17597.832563.922428@notabene.brown> <1176589265.6089.20.camel@heimdal.trondhjem.org> <4622D614.5030500@oracle.com> <1176734503.6761.37.camel@heimdal.trondhjem.org> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Cc: Neil Brown , nfs@lists.sourceforge.net To: Trond Myklebust Return-path: Received: from sc8-sf-mx1-b.sourceforge.net ([10.3.1.91] helo=mail.sourceforge.net) by sc8-sf-list2-new.sourceforge.net with esmtp (Exim 4.43) id 1HdSXO-0007jH-LF for nfs@lists.sourceforge.net; Mon, 16 Apr 2007 07:50:38 -0700 Received: from mx1.redhat.com ([66.187.233.31]) by mail.sourceforge.net with esmtp (Exim 4.44) id 1HdSVd-0001Uk-Ix for nfs@lists.sourceforge.net; Mon, 16 Apr 2007 07:48:49 -0700 In-Reply-To: <1176734503.6761.37.camel@heimdal.trondhjem.org> List-Id: "Discussion of NFS under Linux development, interoperability, and testing." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: nfs-bounces@lists.sourceforge.net Errors-To: nfs-bounces@lists.sourceforge.net Trond Myklebust wrote: > > > On Mon, 16 Apr 2007 at 10:12:09 -0400, Peter Staubach wrote: > >> "llock" or "local lock" implies that only local locks will be used. >> Thus, only the locks that the kernel is managing are used or kept >> track of. No over the wire traffic for locking is issued. >> >> With this option, caching is enabled. >> > > So are you saying that 'llock' basically mirrors our 'nolock', or does > it additionally disable the flushing operations? If the latter, then I > suggest that we add a mount flag, NFS_MOUNT_LLOCK which automatically > sets NFS_MOUNT_NONLM + a new flag that just disables the cache flush on > lock/unlock. I believe that the use of "llock" also disables the cache flush at lock/unlock. Perhaps someone with access to the OpenSolaris code, who isn't worried about CDDL versus GPL issues, could look to verify. It has been a bit since I've worked on that code. I do remember that the "llock" code was designed to be used for things like root file systems, where there shouldn't be any conflicts between different clients, and hence, no need for special page cache handling. Thanx... ps ------------------------------------------------------------------------- This SF.net email is sponsored by DB2 Express Download DB2 Express C - the FREE version of DB2 express and take control of your XML. No limits. Just data. Click to get it now. http://sourceforge.net/powerbar/db2/ _______________________________________________ NFS maillist - NFS@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nfs