From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-ot0-f196.google.com ([74.125.82.196]:42292 "EHLO mail-ot0-f196.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752961AbeDQRrJ (ORCPT ); Tue, 17 Apr 2018 13:47:09 -0400 Received: by mail-ot0-f196.google.com with SMTP id h55-v6so22308479ote.9 for ; Tue, 17 Apr 2018 10:47:08 -0700 (PDT) MIME-Version: 1.0 In-Reply-To: References: <20180417144059.nwbbynhgq3k3i63q@XZHOUW.usersys.redhat.com> <20180417161056.GA24257@infradead.org> <20180417165756.GC24738@magnolia> From: Dan Williams Date: Tue, 17 Apr 2018 10:47:08 -0700 Message-ID: Subject: Re: ioctl FIBMAP for dax gone in v4.17-rc1 Content-Type: text/plain; charset="UTF-8" Sender: linux-xfs-owner@vger.kernel.org List-ID: List-Id: xfs To: Andreas Dilger Cc: "Darrick J. Wong" , Christoph Hellwig , Xiong Zhou , linux-fsdevel , linux-xfs , Eric Sandeen , fstests , linux-nvdimm On Tue, Apr 17, 2018 at 10:40 AM, Andreas Dilger wrote: > On Apr 17, 2018, at 10:57 AM, Darrick J. Wong wrote: >> >> On Tue, Apr 17, 2018 at 09:53:47AM -0700, Dan Williams wrote: >>> On Tue, Apr 17, 2018 at 9:10 AM, Christoph Hellwig wrote: >>>> On Tue, Apr 17, 2018 at 10:40:59PM +0800, Xiong Zhou wrote: >>>>> We got these in v4.17-rc1: >>>>> 6e2608d xfs, dax: introduce xfs_dax_aops >>>>> fb094c9 ext2, dax: introduce ext2_dax_aops >>>>> 5f0663b ext4, dax: introduce ext4_dax_aops >>>>> >>>>> And we don't have ->bmap call in these aops, which may lead >>>>> to the ioctl call failure. >>>>> >>>>> Do we have any plan of adding/supporting it ? >>>>> >>>>> xfstests generic/223 covers this issue. If we are not going >>>>> to support this call for dax, we need to fix the testcase. >>>> >>>> Not supporting ->bmap is a good thing as it is hightly dangerous. >>> >>> I take this to mean "don't fix, it is another casualty of dax being >>> experimental and it won't be coming back". I can get on board with >>> that. >>> >>> Otherwise, I was about to send a series adding bmap to {xfs,ext2,ext4}_dax_ops. >> >> Frankly I'd rather see the swapfile code learn how to iomap and then we >> can get rid of bmap in xfs entirely. > > Is anyone still using LILO to boot? It needed FIBMAP support to map the > kernel image for booting. I don't know if Grub needs FIBMAP support for > the early boot stages or not (it has minimal filesystem support in the > later stages), but it would be a shame if it wasn't possible to boot an > all-NVRAM system as a result of a missing ->bmap() method. Alternately, > convince the Grub folks to use FIEMAP if that is available... For boot the recommendation is to use the BTT (block-translation-table) to provide a sector-atomic block device for the system-partition. Such a configuration does not support DAX mode operation, so there should be no conflict. The EFI system partition is also FAT in most cases which does not support DAX.