From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Bartlomiej Zolnierkiewicz" Subject: Re: [PATCH 0/7] ide: locking improvements Date: Thu, 9 Oct 2008 10:36:14 +0200 Message-ID: <58cb370e0810090136t327734cerf6eb933422b9791c@mail.gmail.com> References: <20081008202930.19112.90371.sendpatchset@localhost.localdomain> <20081009065141.GV19428@kernel.dk> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-path: Received: from rv-out-0506.google.com ([209.85.198.225]:13654 "EHLO rv-out-0506.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751250AbYJIIgP (ORCPT ); Thu, 9 Oct 2008 04:36:15 -0400 Received: by rv-out-0506.google.com with SMTP id k40so4708154rvb.1 for ; Thu, 09 Oct 2008 01:36:15 -0700 (PDT) In-Reply-To: <20081009065141.GV19428@kernel.dk> Content-Disposition: inline Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: Jens Axboe Cc: linux-ide@vger.kernel.org, linux-kernel@vger.kernel.org On Thu, Oct 9, 2008 at 8:51 AM, Jens Axboe wrote: > On Wed, Oct 08 2008, Bartlomiej Zolnierkiewicz wrote: >> >> Locking improvements in preparation for replacing the global ide_lock >> spinlock by per-hwgroup spinlocks [1]. >> >> [1] patch (which is partially based on 2005 patch from Scalex86) for this >> is also ready but it needs some more audit and testing >> >> diffstat: >> drivers/ide/ide-cd.c | 38 ++++++------- >> drivers/ide/ide-io.c | 129 ++++++++++++++++++++--------------------------- >> drivers/ide/ide-ioctls.c | 3 - >> drivers/ide/ide-lib.c | 7 -- >> drivers/ide/ide-proc.c | 25 +-------- >> drivers/ide/ide.c | 7 -- >> 6 files changed, 80 insertions(+), 129 deletions(-) > > Sorry, but I just have to ask 'why'? IDE is seeing a whole lot of churn > for something that should essentially be a stable code base in > maintenance mode, and now scalability improvements? It is the stable code but being in "maintenance only mode" has never been true and as long as there are active users & developers there is really no reason to change it. > Just doesn't make ANY sense to me, sorry. We may end up with a cleaner > code base, but likely also a buggier one. It's not like hardware > coverage testing is all that great, considering some of the ancient > stuff it supports :-) The changes above are relatively safe/simple and are not hardware specific. Thanks for worring about IDE but we should be fine. :) Bart