Return-Path: Received: from quartz.orcorp.ca ([184.70.90.242]:52780 "EHLO quartz.orcorp.ca" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932448AbcFHRk0 (ORCPT ); Wed, 8 Jun 2016 13:40:26 -0400 Date: Wed, 8 Jun 2016 11:40:14 -0600 From: Jason Gunthorpe To: Trond Myklebust Cc: Tom Talpey , Chuck Lever , linux-rdma , Linux NFS Mailing List Subject: Re: [PATCH v1 09/20] xprtrdma: Limit the number of rpcrdma_mws Message-ID: <20160608174014.GB30822@obsidianresearch.com> References: <20160607194001.18401.88592.stgit@manet.1015granger.net> <20160607194732.18401.71941.stgit@manet.1015granger.net> <20160607204941.GA7991@obsidianresearch.com> <20160607212831.GA9192@obsidianresearch.com> <20160607220107.GA9982@obsidianresearch.com> <8FF9BF11-C5BF-4418-9A84-AAD7A6D6321D@primarydata.com> MIME-Version: 1.0 Content-Type: text/plain; charset=unknown-8bit In-Reply-To: <8FF9BF11-C5BF-4418-9A84-AAD7A6D6321D@primarydata.com> Sender: linux-nfs-owner@vger.kernel.org List-ID: On Wed, Jun 08, 2016 at 03:06:56PM +0000, Trond Myklebust wrote: > That would be a mount parameter which would only apply to RDMA since > everything else uses dynamic slot allocation these days. I?d prefer to > avoid that. The only reason RDMA cannot allocate MRs on the fly is performance.. Maybe a dynamic scheme is the right answer.. Jason