From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S933061Ab3BLLiw (ORCPT ); Tue, 12 Feb 2013 06:38:52 -0500 Received: from moutng.kundenserver.de ([212.227.126.171]:51599 "EHLO moutng.kundenserver.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932877Ab3BLLiv (ORCPT ); Tue, 12 Feb 2013 06:38:51 -0500 From: Arnd Bergmann To: Alexander Shiyan Subject: Re: Re[2]: [PATCH v2 1/3] mfd: syscon: Removed support for unloading Date: Tue, 12 Feb 2013 11:38:48 +0000 User-Agent: KMail/1.12.2 (Linux/3.8.0-5-generic; KDE/4.3.2; x86_64; ; ) Cc: linux-kernel@vger.kernel.org, Dong Aisheng , Samuel Ortiz , Mark Brown References: <1360608173-12907-1-git-send-email-shc_work@mail.ru> <201302112005.35034.arnd@arndb.de> <1360650935.475602112@f368.mail.ru> In-Reply-To: <1360650935.475602112@f368.mail.ru> MIME-Version: 1.0 Content-Type: Text/Plain; charset="utf-8" Content-Transfer-Encoding: 7bit Message-Id: <201302121138.48288.arnd@arndb.de> X-Provags-ID: V02:K0:iufxISrvGRUo61NQ723WlchuQYsQmC6pcMOIrXKMD7U wgiSt6sJNxbcuk4tsEJoLNX6RZMsKAzypAQxh/Y55ao13/ABiF hhWURnL518K0L7/sXYqmVumcn8/8VztuMOBItMB75qJUlB+bn2 f571hs9mP10EOjCsPHtQeWvP2+OWdDZkYQQFaSCa/i/ffmcpWg zpIoh3SZdFXf2lfgRlNXst1VFSMSIkFvbHKi9G5amDlEqiBgBr r9sEYca6GoLb4dN5Qko2ex3ArIBgP2Za7J9oPnGgiesf8aArYz tleuQnenEAoRkbug0ggX/euWzEPWeqsKGJ/YYgTOHdGxVMIKQE X/oWbRzEChFBV2vtGWQg= Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tuesday 12 February 2013, Alexander Shiyan wrote: > > On Monday 11 February 2013, Alexander Shiyan wrote: > > > The driver can be used in various subsystems and therefore should not > > > be unloaded when it is defined in the kernel configuration, so remove > > > support for unloading it. > > > > > > Signed-off-by: Alexander Shiyan > > > > Can you describe a scenario where that would happen? Normally the > > module should stay pinned as long as any other module refers > > to its exported symbols. > > Probably I wrote a bad description. > The driver registered by "postcore_initcall". Therefore, if we unregister the > driver, we have no way to register it back. Fixme please. Ah, the driver is actually "bool" in Kconfig rather than "tristate", so it is not possible to build it as a module. Makes sense then, and if we ever want to turn this into a module, we can still revert your patch. Arnd