From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755917Ab3A2CHr (ORCPT ); Mon, 28 Jan 2013 21:07:47 -0500 Received: from szxga02-in.huawei.com ([119.145.14.65]:1311 "EHLO szxga02-in.huawei.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751728Ab3A2CHn (ORCPT ); Mon, 28 Jan 2013 21:07:43 -0500 Message-ID: <51072EDA.2000704@huawei.com> Date: Tue, 29 Jan 2013 10:07:22 +0800 From: Jiang Liu User-Agent: Mozilla/5.0 (Windows NT 5.1; rv:9.0) Gecko/20111222 Thunderbird/9.0.1 MIME-Version: 1.0 To: Yinghai Lu CC: Bjorn Helgaas , "Rafael J. Wysocki" , Jiang Liu , Kenji Kaneshige , Yijing Wang , , , Greg Kroah-Hartman , ACPI Devel Maling List , Toshi Kani , Myron Stowe Subject: Re: [RFC PATCH v5 4/8] ACPI, PCI: avoid building pci_slot as module References: <1358525267-14268-1-git-send-email-jiang.liu@huawei.com> <1358525267-14268-5-git-send-email-jiang.liu@huawei.com> <3500982.MUDp6kKU3R@vostro.rjw.lan> In-Reply-To: Content-Type: text/plain; charset="ISO-8859-1" Content-Transfer-Encoding: 7bit X-Originating-IP: [10.108.108.229] X-CFilter-Loop: Reflected Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On 2013-1-29 6:58, Yinghai Lu wrote: > On Mon, Jan 28, 2013 at 2:14 PM, Bjorn Helgaas wrote: >> >> CONFIG_ACPI_PCI_SLOT=y in RHEL6, so evidently they have this problem. > > oh, I only checked opensuse that has that set to m. > >> >> Asking users to edit module.conf by hand is not a solution, just like >> asking users to boot with a command line option is not a solution. >> That sort of stuff is fine for a hobbyist OS intended only for techie >> geeks. It's not fine for Linux. > > not sure. add something in command line or conf files. > but recompile kernel is another story. > >> >> If you would give a concrete example of the ACPI namespace info and >> device config, hotplug sequence, etc., required to show the problem, >> we could have a useful discussion about ways to fix it. But if all >> you have is FUD about "this might break and users won't have the >> ability to edit modules.conf," that doesn't help me see why this patch >> is a bad idea. > > Never mind, We should save your bandwidth to more patches. Hi Yinghai, Could we use quirk to auto-disable PCIe native hotplug for problematic platforms? Regards! Gerry > > Yinghai > > . >