From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1030980AbXDWF5z (ORCPT ); Mon, 23 Apr 2007 01:57:55 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1030994AbXDWF5z (ORCPT ); Mon, 23 Apr 2007 01:57:55 -0400 Received: from [70.254.190.220] ([70.254.190.220]:41635 "EHLO server.willdawg" rhost-flags-FAIL-FAIL-OK-FAIL) by vger.kernel.org with ESMTP id S1030980AbXDWF5y (ORCPT ); Mon, 23 Apr 2007 01:57:54 -0400 Date: Mon, 23 Apr 2007 05:57:53 +0000 (GMT) From: William Heimbigner X-X-Sender: icxcnika@server.thyself To: Rik van Riel cc: Eric Hopper , linux-kernel@vger.kernel.org Subject: Re: Question about Reiser4 In-Reply-To: <462C4858.3050006@redhat.com> Message-ID: References: <20070423020046.GA28477@omnifarious.org> <462C2E5B.1080008@redhat.com> <462C4858.3050006@redhat.com> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org > William Heimbigner wrote: >> > Eric Hopper wrote: >> > > I know that this whole effort has been put in disarray by the >> > > prosecution of Hans Reiser, but I'm curious as to its status. >> > >> > It was in disarray well before. Many of the reiser4 features, >> > like filesystem plugins, make more technical sense in the Linux >> > VFS, but made more business sense for Namesys as a reiserfs 4 >> > thing. That lead to a stalemate. >> > >> Shouldn't it be a matter of stability though? > > A lot of other things matter. Things like a willingness to > maintain the code after it gets merged, or at least turning > the code into something the community is willing to maintain > if the original developers stop maintaining it. > >> Benchmarks suggest that reiser4 is a good file system; reiser4 is the >> successor to the already-accepted reiserfs; we've got experimental ext4 >> support but no reiser4 support, etc. > > Namesys kind of abandoned reiserfs after work on reiser4 > started. Taking in a new code base on such a track record > is not a good idea when the code is not in a shape where > the community wants to maintain it. > >> I don't see why something like plugins should matter. If it works enough >> to be marked as experimental, why shouldn't reiser4 support be included? >> It's a pain for me personally to have to patch any kernel with reiser4 >> support so I can use the reiser4 fs. > > You basically have three options: > > 1) keep patching every time you upgrade the kernel > > 2) use another filesystem > > 3) become the new reiser4 maintainer and turn the code > into something that Linus is willing to accept I suppose. I have a feeling there's an underlying issue behind "code standards" (and even then, I think that code standards is ultimately an excuse for not integrating reiser4 support into the kernel, but that's just my opinion). However, is the code really in such a shape that the community doesn't want to maintain it? Obviously there's a significant number of people interested in reiser4 - if there weren't, questions like this wouldn't keep getting asked. William Heimbigner icxcnika@mar.tar.cc