From mboxrd@z Thu Jan 1 00:00:00 1970 From: Matthew Garrett Date: Thu, 19 Nov 2009 14:30:08 +0000 Subject: Re: [PATCH] [RFC] Add support for uevents on block device idle Message-Id: <20091119143008.GA719@srcf.ucam.org> List-Id: References: <20091118195342.GA13627@srcf.ucam.org> <20091118200712.GA14026@srcf.ucam.org> <20091118213355.GA16630@srcf.ucam.org> <20091119130107.GB20949@srcf.ucam.org> <20091119141634.GA311@srcf.ucam.org> In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Kay Sievers Cc: David Zeuthen , linux-kernel@vger.kernel.org, axboe@kernel.dk, linux-hotplug@vger.kernel.org On Thu, Nov 19, 2009 at 03:25:59PM +0100, Kay Sievers wrote: > On Thu, Nov 19, 2009 at 15:16, Matthew Garrett wrote: > > It could, but it seems a bit of a hack. It'd still also require the > > timer to be in the kernel, so we might as well expose that to userspace. > > Sure, but a userspace configurable policy for an in-kernel disk-idle > powermanagent sounds fine, compared to a single-subscriber > userspace-only disk-idle event interface. :) Well, we still need to expose this for the access pattern modifying. I really don't see the issue with the single subscriber being devkit-disks - none of the operations involved are atomic, so we're inherently racy here. -- Matthew Garrett | mjg59@srcf.ucam.org