From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S933652Ab3FRSWw (ORCPT ); Tue, 18 Jun 2013 14:22:52 -0400 Received: from h1446028.stratoserver.net ([85.214.92.142]:33602 "EHLO mail.ahsoftware.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932189Ab3FRSWv (ORCPT ); Tue, 18 Jun 2013 14:22:51 -0400 Message-ID: <51C0A560.2020702@ahsoftware.de> Date: Tue, 18 Jun 2013 20:22:24 +0200 From: Alexander Holler User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/20130514 Thunderbird/17.0.6 MIME-Version: 1.0 To: rtc-linux@googlegroups.com CC: Jingoo Han , linux-kernel@vger.kernel.org, "'Andrew Morton'" , "'John Stultz'" , Sachin Kamat , "'Dmitry Torokhov'" Subject: Re: [rtc-linux] Re: BUG: rtc: rtc_device_unregister() broken in 3.10-rcN References: <51BB50B1.7060308@ahsoftware.de> <51BF2DD8.5030004@ahsoftware.de> <002901ce6bc7$71d32510$55796f30$@samsung.com> In-Reply-To: <002901ce6bc7$71d32510$55796f30$@samsung.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Am 18.06.2013 03:59, schrieb Jingoo Han: > On Tuesday, June 18, 2013 12:40 AM, Alexander Holler wrote: >> Am 14.06.2013 19:19, schrieb Alexander Holler: >>> Hello, >>> >>> I've just tested some things with 3.10-rc5+ and discovered that >>> rtc_device_unregister is broken. This is most likely because of the >>> switch to devm*. As 3.10 is already at -rc5, I'm posting this without > > CC'ed Dmitry Torokhov, Sachin Kamat > > > Hi Alexander, > > You are using 'drivers/rtc/rtc-hid-sensor-time.c', right? > If so, would you revert '479243b rtc: hid-sensor-time: use devm_rtc_device_register()', > and give the result? > > > Hi Dmitry, Sachin, > You can see the kernel debug message, when the problem happens. > (http://lists-archives.com/linux-kernel/27855385-bug-rtc-rtc_unregister_device-broken-in-3-10-rcn.html) > I've found the problem. Somehow I managed it with a patch i posted recently (3/9) to get the rtc_device_unregister() back into the remove-function of rtc-hid-sensor-time. Might be happend when rebasing the patch from 3.9.x to 3.10 or similiar, I don't know or remember. I haven't seen it, because the backtrace doesn't show it. So what happend was likely: remove -> rtc_device_unregister() (ok) device remove -> devm -> rtc_device_unregister() (again) -> Crash. Sorry for the noise. Regards, Alexander Holler