From: Jan Kara Subject: Re: [PATCH 1/1] fs/mpage.c: forgotten WRITE_SYNC in case of data integrity write Date: Fri, 14 Mar 2014 15:15:11 +0100 Message-ID: <20140314141511.GF18583@quack.suse.cz> References: <1392519268-29991-1-git-send-email-r.peniaev@gmail.com> <20140218155945.ee2e22e07fa3b4f242b16a7f@linux-foundation.org> <20140313200119.GB504@quack.suse.cz> <20140313143456.157404fd7f208638ca70e317@linux-foundation.org> <20140314130712.GC12613@htj.dyndns.org> <20140314141143.GH12613@htj.dyndns.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Roman Peniaev , Andrew Morton , Jan Kara , Alexander Viro , linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, Jens Axboe To: Tejun Heo Return-path: Content-Disposition: inline In-Reply-To: <20140314141143.GH12613@htj.dyndns.org> Sender: linux-kernel-owner@vger.kernel.org List-Id: linux-fsdevel.vger.kernel.org On Fri 14-03-14 10:11:43, Tejun Heo wrote: > > Also, could you please help me do understand how can I guarantee > > integrity in case of block device with big volatile > > cache and filesystem, which does not support REQ_FLUSH/FUA? > > If a device has a volatile cache but doesn't support flush, it can't > guarantee integrity. There's no way for its user to determine or > force whether certain data is on non-volatile media. It's an > inherently broken device. I think his problem was that the device does support REQ_FLUSH/FUA but the filesystem on top of it doesn't issue it properly. That's a filesystem problem so fix the filesystem... :) Which one is it? Honza -- Jan Kara SUSE Labs, CR