From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1761642AbdEVUxL (ORCPT ); Mon, 22 May 2017 16:53:11 -0400 Received: from mga01.intel.com ([192.55.52.88]:12690 "EHLO mga01.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752436AbdEVUxK (ORCPT ); Mon, 22 May 2017 16:53:10 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.38,379,1491289200"; d="scan'208";a="971792908" Date: Mon, 22 May 2017 23:48:36 +0300 From: Mika Westerberg To: Mario.Limonciello@dell.com Cc: gregkh@linuxfoundation.org, andreas.noever@gmail.com, michael.jamet@intel.com, yehezkel.bernat@intel.com, lukas@wunner.de, amir.jer.levy@intel.com, luto@kernel.org, Jared.Dominguez@dell.com, andriy.shevchenko@linux.intel.com, linux-kernel@vger.kernel.org Subject: Re: [PATCH 00/24] Thunderbolt security levels and NVM firmware upgrade Message-ID: <20170522204836.GH8541@lahna.fi.intel.com> References: <20170518143914.60902-1-mika.westerberg@linux.intel.com> <799e3a9d4d394d1eb9b9e742d81bd044@ausx13mpc120.AMER.DELL.COM> <20170519171948.GG8541@lahna.fi.intel.com> <20170520082412.GI8541@lahna.fi.intel.com> <20170522113709.GX8541@lahna.fi.intel.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: Organization: Intel Finland Oy - BIC 0357606-4 - Westendinkatu 7, 02160 Espoo User-Agent: Mutt/1.7.1 (2016-10-04) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, May 22, 2017 at 08:07:54PM +0000, Mario.Limonciello@dell.com wrote: > I was 1 version behind, but I double checked with the latest version (1.1.15) > and the same behavior exists on Linux (still works properly on Win10). > > If you have some more details about what the FW guys changed, I can check > with my Dell FW team if they've picked up the same fix. I'm guessing it's not > the same problem though considering it works properly on Win10? Can you send me full dmesg preferably so that you have acpiphp.dyndbg in the kernel command line (you also need to have CONFIG_DYNAMIC_DEBUG=y)? It is possible that Windows does ACPI hotplug differently, for example it could add some delay somewhere and that is enough for the firmware to get the bridges initialized properly where as in Linux we only see the bridge when it is in the middle of the initialization or so. When it works properly Linux should see all the PCI bridges configured properly by the BIOS SMI handler.