From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1750756AbWGFTlw (ORCPT ); Thu, 6 Jul 2006 15:41:52 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1750760AbWGFTlw (ORCPT ); Thu, 6 Jul 2006 15:41:52 -0400 Received: from cantor.suse.de ([195.135.220.2]:7845 "EHLO mx1.suse.de") by vger.kernel.org with ESMTP id S1750756AbWGFTlv (ORCPT ); Thu, 6 Jul 2006 15:41:51 -0400 Date: Thu, 6 Jul 2006 12:38:04 -0700 From: Greg KH To: Andrew Morton Cc: tiwai@suse.de, arjan@infradead.org, kkeil@suse.de, torvalds@osdl.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH] hisax fix usage of __init* Message-ID: <20060706193804.GB20621@kroah.com> References: <20060705112357.GA7003@pingi.kke.suse.de> <1152099459.3201.19.camel@laptopd505.fenrus.org> <20060706191129.GA20255@kroah.com> <20060706122250.34fdeded.akpm@osdl.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20060706122250.34fdeded.akpm@osdl.org> User-Agent: Mutt/1.5.11 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Thu, Jul 06, 2006 at 12:22:50PM -0700, Andrew Morton wrote: > On Thu, 6 Jul 2006 12:11:29 -0700 > Greg KH wrote: > > > > > Tips: > > > > The module_init()/module_exit() functions (and all initialization > > > > functions called only from these) should be marked __init/exit. > > > > The struct pci_driver shouldn't be marked with any of these tags. > > > > The ID table array should be marked __devinitdata. > > > > Yes, and that is correct. They should never be marked __initdata, as > > that is wrong for when CONFIG_HOTPLUG is enabled and the module is built > > in. > > > > So either use __devinitdata, or nothing (as it's only a memory savings > > if CONFIG_HOTPLUG is not enabled, which is real tough these days, and > > the driver is built into the system.) > > I think the problem is that pci_driver has a pointer to the id_table. So > we have a ref to __devinitdata from .text. > > That's runtimely-correct, but the new section checker could get offended. Yes, it probably could get hard to check for it, but the code is correct :) thanks, greg k-h