From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Received: from mout.perfora.net ([74.208.4.194]:65455 "EHLO mout.perfora.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753419Ab0DUCLD (ORCPT ); Tue, 20 Apr 2010 22:11:03 -0400 Message-ID: <4BCE5EAD.4070107@vorgon.com> Date: Tue, 20 Apr 2010 19:10:53 -0700 From: "Timothy D. Lenz" MIME-Version: 1.0 To: linux-media@vger.kernel.org Subject: Re: cx5000 default auto sleep mode References: <4BC5FB77.2020303@vorgon.com> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-media-owner@vger.kernel.org List-ID: On 4/14/2010 10:40 AM, Devin Heitmueller wrote: > On Wed, Apr 14, 2010 at 1:29 PM, Timothy D. Lenz wrote: >> Thanks to Andy Walls, found out why I kept loosing 1 tuner on a FusionHD7 >> Dual express. Didn't know linux supported an auto sleep mode on the tuner >> chips and that it defaulted to on. Seems like it would be better to default >> to off. If someone wants an auto power down/sleep mode and their software >> supports it, then let the program activate it. Seems people are more likely >> to want the tuners to stay on then keep shutting down. >> >> Spent over a year trying to figure out why vdr would loose control of 1 of >> the dual tuners when the atscepg pluging was used thinking it was a problem >> with the plugin. > > The xc5000 power management changes I made were actually pretty > thoroughly tested with that card (between myself and Michael Krufky, > we tested it with just about every card that uses the tuner). In > fact, we uncovered several power management bugs in other drivers as a > result of that effort. It was a grueling effort that I spent almost > three months working on. > > Generally I agree with the premise that functionality like this should > only be enabled for boards it was tested with. However, in this case > it actually was pretty extensively tested with all the cards in > question (including this one), and thus it was deemed safe to enable > by default. We've had cases in the past where developers exercised > poor judgement and blindly turned on power management to make it work > with one card, disregarding the possible breakage that could occur > with other cards that use the same driver -- this was *not* one of > those cases. > > If there is a bug, it should be pretty straightforward to fix provided > it can be reproduced. > > Regarding the general assertion that the power management should be > disabled by default, I disagree. The power savings is considerable, > the time to bring the tuner out of sleep is negligible, and it's > generally good policy. > > Andy, do you have any actual details regarding the nature of the problem? > > Devin > Went down a second time today, so copied the logs over again. If what is needed to fix this isn't in those, will have to look else where.