Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756482AbcCQNtV (ORCPT ); Thu, 17 Mar 2016 09:49:21 -0400 Received: from mx1.redhat.com ([209.132.183.28]:39254 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752083AbcCQNtS (ORCPT ); Thu, 17 Mar 2016 09:49:18 -0400 Subject: Re: [PATCH 2/2] block: create ioctl to discard-or-zeroout a range of blocks To: Chris Mason , sandeen@redhat.com, Linus Torvalds , Dave Chinner , "Theodore Ts'o" , Ric Wheeler , Andy Lutomirski , One Thousand Gnomes , Gregory Farnum , "Martin K. Petersen" , Christoph Hellwig , "Darrick J. Wong" , Jens Axboe , Andrew Morton , Linux API , Linux Kernel Mailing List , shane.seymour@hpe.com, Bruce Fields , linux-fsdevel , Jeff Layton References: <56E69398.7030508@redhat.com> <20160314144603.GO29218@thunk.org> <20160315201431.GG30721@dastard> <20160315223313.GH30721@dastard> <20160315235216.GI30721@dastard> <56E8A916.8050702@redhat.com> <20160316005117.GA34410@clm-mbp.thefacebook.com> <20160316222343.GA53649@clm-mbp.thefacebook.com> From: Ric Wheeler Message-ID: <56EAB5DB.8080106@redhat.com> Date: Thu, 17 Mar 2016 09:49:15 -0400 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20100101 Thunderbird/38.5.0 MIME-Version: 1.0 In-Reply-To: <20160316222343.GA53649@clm-mbp.thefacebook.com> Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2697 Lines: 65 On 03/16/2016 06:23 PM, Chris Mason wrote: > On Tue, Mar 15, 2016 at 05:51:17PM -0700, Chris Mason wrote: >> On Tue, Mar 15, 2016 at 07:30:14PM -0500, Eric Sandeen wrote: >>> On 3/15/16 7:06 PM, Linus Torvalds wrote: >>>> On Tue, Mar 15, 2016 at 4:52 PM, Dave Chinner wrote: >>>>>> It is pretty clear that the onus is on the patch submitter to >>>>>> provide justification for inclusion, not for the reviewer/Maintainer >>>>>> to have to prove that the solution is unworkable. >>>> I agree, but quite frankly, performance is a good justification. >>>> >>>> So if Ted can give performance numbers, that's justification enough. >>>> We've certainly taken changes with less. >>> I've been away from ext4 for a while, so I'm really not on top of the >>> mechanics of the underlying problem at the moment. >>> >>> But I would say that in addition to numbers showing that ext4 has trouble >>> with unwritten extent conversion, we should have an explanation of >>> why it can't be solved in a way that doesn't open up these concerns. >>> >>> XFS certainly has different mechanisms, but is the demonstrated workload >>> problematic on XFS (or btrfs) as well? If not, can ext4 adopt any of the >>> solutions that make the workload perform better on other filesystems? >> When I've benchmarked this in the past, doing small random buffered writes >> into an preallocated extent was dramatically (3x or more) slower on xfs >> than doing them into a fully written extent. That was two years ago, >> but I can redo it. > So I re-ran some benchmarks, with 4K O_DIRECT random ios on nvme (4.5 > kernel). This is O_DIRECT without O_SYNC. I don't think xfs will do > commits for each IO into the prealloc file? O_SYNC makes it much > slower, so hopefully I've got this right. > > The test runs for 60 seconds, and I used an iodepth of 4: > > prealloc file: 32,000 iops > overwrite: 121,000 iops > > If I bump the iodepth up to 512: > > prealloc file: 33,000 iops > overwrite: 279,000 iops > > For streaming writes, XFS converts prealloc to written much better when > the IO isn't random. You can start seeing the difference at 16K > sequential O_DIRECT writes, but really its not a huge impact. The worst > case is 4K: > > prealloc file: 227MB/s > overwrite: 340MB/s > > I can't think of sequential workloads where this will matter, since they > will either end up with bigger IO or the performance impact won't get > noticed. > > -chris I think that these numbers are the interesting ones, see a 4x slow down is certainly significant. If you do the same patch after hacking XFS preallocation as Dave suggested with xfs_db, do we get most of the performance back? Ric