From mboxrd@z Thu Jan 1 00:00:00 1970 From: Davidlohr Bueso Subject: Re: [PATCH 3/6] drivers/IB,qib: do not use mmap_sem Date: Tue, 29 Jan 2019 06:14:45 -0800 Message-ID: <20190129141445.kch4jxr2ps62ohcw@linux-r8p5> References: <20190121174220.10583-1-dave@stgolabs.net> <20190121174220.10583-4-dave@stgolabs.net> <20190128233140.GA12530@ziepe.ca> <20190129044607.GL25106@ziepe.ca> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii; format=flowed Return-path: Content-Disposition: inline In-Reply-To: <20190129044607.GL25106@ziepe.ca> Sender: linux-kernel-owner@vger.kernel.org To: Jason Gunthorpe Cc: akpm@linux-foundation.org, dledford@redhat.com, jack@suse.de, ira.weiny@intel.com, linux-rdma@vger.kernel.org, linux-mm@kvack.org, linux-kernel@vger.kernel.org, dennis.dalessandro@intel.com, mike.marciniszyn@intel.com, Davidlohr Bueso List-Id: linux-rdma@vger.kernel.org On Mon, 28 Jan 2019, Jason Gunthorpe wrote: >.. and I'm looking at some of the other conversions here.. *most >likely* any caller that is manipulating rlimit for get_user_pages >should really be calling get_user_pages_longterm, so they should not >be converted to use _fast? Yeah this was something that crossed my mind could come up when I first sent the series. I'll send a v3 where I leave the gup() calls alone for rdma drivers; and we can at least convert the mmap_sem to reader. Thanks, Davidlohr