Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753818AbdF0U6h (ORCPT ); Tue, 27 Jun 2017 16:58:37 -0400 Received: from aserp1040.oracle.com ([141.146.126.69]:23416 "EHLO aserp1040.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753437AbdF0U6a (ORCPT ); Tue, 27 Jun 2017 16:58:30 -0400 Subject: Re: [PATCH RFC] hugetlbfs 'noautofill' mount option To: Andrea Arcangeli References: <326e38dd-b4a8-e0ca-6ff7-af60e8045c74@oracle.com> <7ff6fb32-7d16-af4f-d9d5-698ab7e9e14b@intel.com> <03127895-3c5a-5182-82de-3baa3116749e@oracle.com> <22557bf3-14bb-de02-7b1b-a79873c583f1@intel.com> <7677d20e-5d53-1fb7-5dac-425edda70b7b@oracle.com> <48a544c4-61b3-acaf-0386-649f073602b6@intel.com> <476ea1b6-36d1-bc86-fa99-b727e3c2650d@oracle.com> <20170509085825.GB32555@infradead.org> <1031e0d4-cdbb-db8b-dae7-7c733921e20e@oracle.com> <20170616131554.GD11676@redhat.com> <47ea78b4-3b14-264e-2c92-e5e507fd3cba@oracle.com> Cc: Christoph Hellwig , Dave Hansen , linux-kernel@vger.kernel.org, linux-mm@kvack.org, Mike Rapoport , Mike Kravetz From: Prakash Sangappa Message-ID: Date: Tue, 27 Jun 2017 13:57:48 -0700 User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.11; rv:45.0) Gecko/20100101 Thunderbird/45.8.0 MIME-Version: 1.0 In-Reply-To: <47ea78b4-3b14-264e-2c92-e5e507fd3cba@oracle.com> Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 7bit X-Source-IP: userv0022.oracle.com [156.151.31.74] Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 783 Lines: 24 On 6/20/17 4:35 PM, Prakash Sangappa wrote: > > > On 6/16/17 6:15 AM, Andrea Arcangeli wrote: >> Adding a single if (ctx->feature & UFFD_FEATURE_SIGBUS) goto out, >> branch for this corner case to handle_userfault() isn't great and the >> hugetlbfs mount option is absolutely zero cost to the handle_userfault >> which is primarily why I'm not against it.. although it's not going to >> be measurable so it would be ok also to add such feature. > > > If implementing UFFD_FEATURE_SIGBUS is preferred instead of the mount > option, I could look into that. > Implementing UFFD_FEATURE_SIGBUS seems reasonable. I wanted to note here on this thread that I sent out a seperate RFC patch review for adding UFFD_FEATURE_SIGBUS. See, http://marc.info/?l=linux-mm&m=149857975906880&w=2