From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Received: from mail.active-venture.com ([67.228.131.205]:58187 "EHLO mail.active-venture.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756451Ab3LECqz (ORCPT ); Wed, 4 Dec 2013 21:46:55 -0500 Message-ID: <529FE919.9060604@roeck-us.net> Date: Wed, 04 Dec 2013 18:46:49 -0800 From: Guenter Roeck MIME-Version: 1.0 To: One Thousand Gnomes CC: wim@iguana.be, linux-watchdog@vger.kernel.org Subject: Re: [PATCH] sc1200_wdt: Fix oops References: <20131204153135.32175.35874.stgit@alan.etchedpixels.co.uk> <20131204163118.GA1299@roeck-us.net> <20131205001511.6c8e9f60@alan.etchedpixels.co.uk> In-Reply-To: <20131205001511.6c8e9f60@alan.etchedpixels.co.uk> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-watchdog-owner@vger.kernel.org List-Id: linux-watchdog@vger.kernel.org On 12/04/2013 04:15 PM, One Thousand Gnomes wrote: > On Wed, 4 Dec 2013 08:31:18 -0800 > Guenter Roeck wrote: > >> On Wed, Dec 04, 2013 at 03:31:52PM +0000, Alan wrote: >>> If loaded with isapnp = 0 the driver explodes. This is catching >>> people out now and then. What should happen in the working case is >>> a complete mystery and the code appears terminally confused, but we >> >> Looks like it uses the pnp infrastructure only to auto-detect the >> "io" address, and assumes that pnp_register_driver() calls the >> pnp probe function immediately. Afterwards it unregisters and >> works (or is supposed to work) outside the pnp subsystem. Brr. >> >>> can at least make the error path work properly. >>> >>> Signed-off-by: Alan Cox >>> Partially-Resolves-bug: https://bugzilla.kernel.org/show_bug.cgi?id=53991 >> >> Actually, I think it fully resolves it, because the other part (sc520 driver >> not working on SC1100 CPU) isn't a bug. It isn't working on an ARM or PPC CPU >> either, nor on any Intel CPU. > > As far as I can tell the fact it fails on the sc1100 is also a bug - but > the mess of idefs around PNP is impossible to figure out without some > actual docs, time and real hardware. > ... unless isapnp was set to 0 via module parameter, or if SC1100 is not compatible with SC1200. But I agree, hard to say w/o datasheet and hw. Either case, seems to me the driver is screaming for a substantial overhaul. Guenter